Hey everyone, welcome back to Academia Obscura. I'm your host, Chris. So a couple weeks ago I did an episode where I put a bunch of blinking red lights on the whiteboard and walked through all of them. Energy, the AI bubble, the debt, shadow banking, the Fed being stuck. And I said at the time, I'm not predicting anything, I'm just showing you the board. Today I wanted to do something narrower because four of those lights changed color in the last two weeks. And one thing happened on Friday that I think almost nobody covered properly. Let me start with Friday. Kevin Warsh gave his first Jackson Hole speech as Fed chairman. And he opened it with a joke, which I love, because the joke is the whole episode. He said there are two kinds of hikes you can take in the mountains around Jackson Hole. There's the brutal death marks, the steely marathon version, which he associated with former vice chairman Don Cone, and then there's the leisurely stroll along the easy trails, which he associates with Ben Bernanke. And his advice to the room was quote, ask yourself, is this a Cone day or a Bernanke day? That's the chairman of the Federal Reserve on his hundredth day in the job, making a joke about not being Ben Bernanke in a speech where he told everyone he might raise rates. Last episode I told you Warsh is the anti-Bernanke. On Friday, he got up in Wyoming and said it himself as a punchline. Okay. So here's the line that actually mattered. Quote, we must be confident that underlying inflation is moving to our objective clearly and at sufficient speed. Otherwise, we have work to do. And on the Summer's inflation numbers, which were better than expected, he said they do not tell me that underlying trends have meaningfully improved. Now, I want to be careful here because I've seen people overstate this. He did not announce a rate hike. He didn't even give a clear signal. Reuters called it the closest he's come to acknowledging that hikes might be needed, and I think that's the right framing. But the market heard it and moved hard. Odds of a September rate hike went from somewhere in the mid-30s to about 60% before lunch. As of this morning, Fed funds futures have it at a 57%, and the prediction markets are lower. Polymarkets got the 25 basin point hike at 44. The two-year Treasury jumped 11 basis points to 434, the highest it's been in a month. And here's the part I haven't seen anybody report, and it's the interesting part. The two-year moved 11 basis points. The 30-year moved 1.6. Think about what that means. The front end of the curve went haywire and the long end barely twitched. If the market were repricing inflation, the long end moves. If the market were repricing growth, the long end moves. The long end didn't move. So the market wasn't repricing the economy. It was repricing one decision in one room in September. That's about as clean a signal as the market ever gives you. Here's one more thing he said, and honestly, I think it's the most remarkable sentence in the whole speech. And I've seen basically nobody quote it. Quote, the responsibility for 65 months of sustained elevated inflation sits squarely with the central bank. And that is where it belongs. 65 months. That's five and a half years. That's the chairman of the Federal Reserve standing up in public and saying his own institution owns all of it. You don't say that unless you're building a case for doing something about it. And then here's what he didn't say, which I think is just as important. He gave no forward guidance, no reaction function. And that was deliberate. He spent a whole section arguing that forward guidance, in his words, has overstated its welcome. His line was we should not indulge a regime in which market participants are looking primarily to the Fed for their next trade. But here's the absence I actually care about. He did not discuss the standing repo facility. He did not discuss the FEMA line. He did not discuss who was allowed to borrow from the Fed when things get tight. And I went and pulled the transcript off the Fed's website and searched it. The words repo, collateral, liquidity, and financial stability do not appear anywhere in the body of that speech. He talked about artificial intelligence. He talked about how the Fed communicates. He talked about inflation. The plumbing never came up. And that's either because the plumbing is fine or because it's the thing you don't bring up at Jackson Hole. All right, housekeeping. Before I go any further, I got some things wrong last episode and I want to clean them up. I do use AI to help draft and fact check the episodes. If anything ever gets through, I will own that. If you catch anything, please write me and let me know because I'd rather clean it up myself than have somebody do it for me in the comments. One, and this is a bad one. I said crude had crossed $100 a barrel. It hasn't. Brent's currently at $88.10 as of this morning. It got up to $89.70 Thursday. Swiss's been flirting with $90, but it has not crossed $100. Now it's up about 30% on the year, which is a real move, and it's bad enough on its own, but it's not $100. And I'm not going to inflate a number to make my case stronger. If my argument needs oil at $100 to work, then it's a bad argument and you should know that. Two, I said the Fed could just buy stocks to stop a sell-off. It can't. Section 14 of the Federal Reserve AI Act limits what they're allowed to purchase, basically government securities, agency debt, and a couple of other instruments. The Bank of Japan bought stock funds for years. The Fed has no authority to do that. Closest it's ever come was 2020 when it bought corporate bonds and bond ETFs through a special vehicle, backstoped with Treasury money, bonds, not equity, and it needed the Treasury Secretary to sign off. Three, I said the basis trade runs had almost a hundred to one leverage. The Cleveland Fed put the median around 17.5, the average around 21. The tail goes much higher. Still terrifying, but I should use the real number. Four, I said inflation was around 4%. CPI came in at 3.4. PCE, the one that Fed actually watches, printed 3.7 headline and 3.3 core on Wednesday. 5. I used BlackRock as a shadow banking example and said they guarantee dividends. They don't. Bad example. The argument works fine without it. Alright, that's housekeeping. Here's what's new. Delta. Four things changed since last we talked. And there's a story from 1932 that I think reframes the entire Fed conversation. One, the market has moved to my Fed call. What I said they might hike, that was a minority view. It is not a minority view anymore. Two, NVIDIA had a blowout quarter on Tuesday, and then on Wednesday they quietly stopped guaranteeing other people's debt. Those two facts do not sit comfortably together. Three, South Korea already had the crash I've been describing. Not a warning sign, a leading indicator. It happened in July, and almost nobody here connected it to us. Four, the United States pulled its last aircraft carrier out of the Western Pacific. Now, before I get into all that, I want to be honest about something. The full chain, all six risks, start to finish, uh, the way I laid it out last time. That's maybe one in ten, probably less. That's according to my chat. I'd rather tell you that in the first five minutes than bury it at the end. The individual pieces are more likely than the whole thing, and that's how I want to be judged. And one more uncomfortable thing. When I first made this Fed call, it was priced around 30%. Back that's when I was researching the episode hard back on Thursday. Futures have it at 57% this morning. The market moved to me while I was building this episode. That's not a victory lap, that's a warning, and here's why. A non-consensus call is worth making. A consensus call is already in the price. So the interesting question stopped being whether the Fed hikes, the interesting question is what the hike lands on. All right, now the chain reaction. So I want to tell you a story, and it's the spine of this whole episode. Everybody reaches for 1929 when they talk about the Great Depression. I want to talk about 1931 because 1931 is the year the machinery actually broke, and the sequence is uncomfortably close to what we're looking at right now. May 11th, 1931. A bank called Credit, uh, sorry, a it is the largest bank in the country, and international lenders start pulling money out of Central Europe. The contagion moves to Germany. Hoover puts together a moratorium in July, a one-year suspension of war debts and repatriations. And then a couple months later, the private bankers add what they called a standstill agreement, international forbearance. Everybody agreed not to call the loans. And it does nothing for domestic funding. Nothing. September 21st, 1931, Britain leaves the gold standard. Now, watch what happens next because this is the part. The New York Fed raises its discount rate from 1.5% to 2.5% on October 9th, and then again to 3.5% on October 16th. Two hikes, one week apart. Not because the American economy needs tightening. The American economy was in the third year of a depression. They did it because gold was flowing out of the country and the currency needed defending. And at the same month, October 1931, 512 American banks closed, tied up $566 million in deposits. It set a monthly record for bank closings. So think about this timeline. A bank fails in Austria in May. Britain leaves the gold on September 21st. The Fed hikes October 9th and October 16th. 512 banks close that same month. The rate decision and the seizure are not five months apart. They're the same three weeks. Now, three things before I go any further because I'm not interested in being a Great Depression crank. And there are real reasons this analogy breaks. One, there is no gold standard. In 1931, the Fed was generally trapped by convertibility. They had to defend gold. It was the law. Today they don't. If work hikes into a shock, that's a choice. And choices can be reversed in an afternoon. Two, there is deposit insurance. The specific mechanism that produced 9,000 bank failures in the 30s cannot repeat the same way. That's what the FDIC is for, and it exists precisely because of 1931. And the banks right now are profitable and well capitalized. If I'm wrong about any of this, that's probably why. So, when I say history rhymes, I don't mean the numbers line up. I mean the sequence rhymes. The shock absorbers are different and they might hold. If I miss this call, I'll miss it because the shock absorbers worked. One more thing about Hoover, then I'll move on. The caricature is that Hoover did nothing, that he sat there and let it happen. That's just false. He ran real deficits, 52.5% and 43.3% of total federal outlays in 1931 and 1932. Okay? He created agencies. He intervened constantly. The problem was never that Hoover did nothing. The problem was that serious people operating inside real constraints acted late and they acted with tools they already decided in advance were acceptable. That's the pattern I'm worried about. Not stupidity, not malice, orthodoxy plus a clock, plus a little bit of stupidity because of course Donald Trump. Energy and Suez. Okay, so let's talk about the fuse, which is energy. Quick facts. The Strategic Petroleum Reserve is at about 209. Sorry, quick facts. The Strategic Petroleum Reserve is at about 299 million barrels against a design capacity of 680. That's the lowest it's been since January 1983. And the GAO found the thing can only draw down and refill at roughly 60 and 65% of its design rates, which means even the barrels that are in, they don't come out as fast as the spec sheet says. The IEA did a 400 million barrel coordinated release earlier this year. That was the one shot. That's spent. Brent is at 88.10, up about 31% year over year, though basically flat over the past month. And I'll give you the honest counter argument, which is that demand is down and 1.6 million barrels a day, and that's real, largely from China, and it's part of why the price hasn't run. Now, here's the thing. There is no Hoover rhyme for an oil checkpoint. 1931 was a monetary crisis. It didn't have one, but 1956 did, and almost nobody tells the story. So in 1956, you've got the Suez crisis. Britain and France and Israel move against Egypt over the canal. Anglo-Fridge air operations start October 31st. The landings are November 5th. And the canal ends up shut to traffic from November 1956 all the way to April of 1957. Britain had the army. Britain took Port Said and the key points in the canal zone. Militarily they won. And then they lost anyway. They lost because the United States declined to support the pound and opposed British access to IMF money and used the fund as leverage to force a withdrawal. Macmillan, who was Chancellor at the time and had been one of the strongest voices for the whole operation, told the cabinet they'd lost $280 million of reserves in a single week. The parliamentary record from December 4th shows reserves falling $279 million down to about $1.97 billion, roughly 12% in a month. And McMillan flipped. The guy who wanted the operation became the guy demanding they pull out. Not because of casualties, because of the reserve position. By December, the British were rationing petrol. That rationing lasted until May. So the invasion starts at the end of October, and the Empire finds out exactly what it is in about six weeks. Now, here's the part that you make you sit up. At Suez, the United States won by controlling somebody else's funding. That was the weapon, not troops, access to dollars. Today, we're the ones who need the funding. There's about 9.3 trillion in treasuries held offshore. Foreign official holdings have been falling on net, down about 70 billion in June alone, according to Treasury's own data. And once you adjust for a measurement gap that the Fed's own staff puts at roughly $1.4 trillion as of the end of 2024, Cayman Islands funds rank as the single largest fold and holder of American government debt. That's a measurement gap, by the way, not a current exposure count. I want to be precise about that. And here's where the analogy breaks, and I'll say it right now. Britain in 1956 was a declining reserve currency with a fixed peg to defend and limited convertibility. The dollar floats. There is no peg, and there's no external authority that can do to us what we did to London. Nobody's going to cut us off from our own currency. That's not the claim. The claim is narrower, but I think it's worse. The country that once won a war by controlling the other side's funding is now the country whose own funding runs through offshore leverage it cannot see. The September Fett. So that's the shock. Now the response. And I want to name what I'm doing here because I'm using two pieces of history tonight and they're doing different jobs. Suez 1956 is the rhyme for the shock. A choke point closes and a great power finds out what it actually depends on. Hoover 1931 is the rhyme for the response. A central bank defends its credibility instead of its economy, and the plumbing ceases about a month later. One is how the fire starts, the other is how the fire department makes it worse. Okay, so where's the Fed right now? The July 29th meeting. They held rates, but the vote was 9-3. Hammock, Kosharki, and Logan each wanted a quarter point hike. And I want to be precise about something here because it matters institutionally. All three of those are regional bank presidents. The governors, the people appointed to the board in Washington, held ranks with Warsh. So it's not that Warsh has lost its board. What he has is a regional insurgency. The minutes from that meeting came out August 19th, and this is the real evidence. Several participants were ready to hike then, and many said tightening would likely be necessary if inflation didn't come down. Many. That's not a fringe, that's most of the root. And then Warsh gets up Friday and says the 2% target is, quote, a firm fixed target. And as that price stability is not self-executing, nor does inflation necessarily mean reverting, which is a very academic way of saying this doesn't fix itself, and it's our job. Now, let me steel man the other side because there's a decent argument that he doesn't need to hike at all. The argument goes, yields have already risen without him doing anything, the long end has backed up, the market did its tightening for him, so why spend the political capital? And here's my answer to that. There are two very different reasons long yields go up and they look identical on a chart. If yields rise because investors is if yields rise because investors expect you to tighten, that's your policy working. That's transmission. If yields rise because investors want more compensation to hold your paper in the first place, that's your bond market getting nervous about you. Same line on the screen, opposite meetings. And I'd argue if you can't tell which one you're looking at, hiking is how you find out. Here's the Hoover tie, and I think this is the center of the whole episode. In 1931, the Fed had a choice between the domestic economy and the credibility of the currency. And they chose credibility. And I want to be fair to them. It was defensible at the time, and the people who made that call were not fools. They were operating inside a constraint they could not see past. October did the rest. Now on Friday, a man who has said in public that he wants to be vulker and not Bernanke stood up at Jackson Hole and joked about not being Bernanke. And wanting to be vulgar is its own kind of gold standard. Think about that. In 1931, the constraint was written down. It was law. Confurtibility. The Fed had no choice. Warsh has no such excuse, and that's exactly why I think this is more worrying, not less. He's choosing the constraint, and a chosen constraint gets defended harder than an imposed one because your reputation is inside it. The Bank of Japan. Alright, Japan. And I'm going to keep this simple because the mechanics get technical fast. The Bank of Japan is at 1%, which is a 31-year high for them. And Reuters is reporting they could go again as soon as their September 17th to 18th meeting. The yin is at a multi-decade lows despite a big intervention earlier this year that did not hold. And if anything, that just gave carry traders a better entry. But here's the piece that I think matters more, and it's not about intervention at all. Picture a Japanese insurance company. It's going to have to pay out claims in yen someday. But for years it's been buying American government bonds because they paid more than Japanese ones. And to protect itself from the exchange rate moving against it, it buys insurance on the currency. Two things just happened at once. That insurance got more expensive, and Japanese government bonds started paying over 4% at the long end. So the guy in Tokyo doesn't panic. He doesn't dump anything. He just doesn't shove up to the next auction. He lets we got run off and then he buys at home instead. For scale, the major Japanese life insurers are sitting on something like 30 trillion yen of unrealised losses on their domestic bonds, called a $194 billion, at as of the end of June. And that's the thing. That's a buyer quietly disappearing, not a seller showing up. You don't see any of it on a screen. You see it on the next auction when the bidding isn't there. In 1931, London moved and New York had to react. The foreign decision became an American funding problem. Today, Tokyo moves and Washington has to react, and it runs through exactly the same place. Who's willing to hold the paper? The hidden leverage. Okay, so who is holding the paper? Increasingly, the answer is hedge funds using borrowed money. The Dallas Fed put out research in May finding that hedge fund net repo borrowing hit roughly $1.8 trillion, which is about 6% of all marketable treasury notes and bonds. As of the end of last year, the basis trade specifically is around $830 billion, leverage in the 10 to 20s on average, with a much fatter tail. And I want to kill a narrative here because I hear it constantly and it's wrong. People say foreigners are dumping treasuries. Foreign official holders, central banks and sovereign funds, they are selling on net. But foreign private holdings are at a record, around $5.5 trillion. So it's not a buyer's strike, it's a handoff. The paper is moving from holders who don't care about price and can sit through anything, to holders who are leveraged and can get margin called. Official money does not get margin called. That's the whole difference. And here's the 1932 parallel, which I love. In 1932, Congress passed a law letting the Fed lend against a wider range of collateral. Section 10B of the Federal Reserve Act, added by the Banking Act of 1932. They passed it as a temporary measure. It was made permanent in 1935. Why'd they need it? Because under the old rules, the system had literally run out of eligible paper. The banks had assets, but not the kind the Fed was allowed to lend against. And that is exactly today's argument about who's allowed to borrow from the standing repo facility. 94 years apart, same bottleneck. When the pipes clog, the fight is always about who gets to be at the window. The one that argues against me. Now I want to give you a number that cuts against everything I'm saying because I think you should have it. The Office of Financial Research publishes a stress test for exactly this kind of thing. Right now it's at negative 2.74, deeply negative, and it's been sitting down there all week. By the United States government's own measure, there's no stress in the system at all. That's either strong evidence that I'm wrong, or it's the same reading you would have got in the summer of 2007. My money is that the government is just misreading all the data. More on that next episode. But like I said, that that is a subjective call. Britain 2022. So let me give you a case where this actually happened. Recently, in a wealthy country with a real central bank. September 2022. The UK government puts out a mini budget. 45 billion pounds of unfunded tax cuts. Sterling drops to an all-time low against the dollar. 132 10327 on September 26th. 30 yield get yields spike 120 basis points in three days. And here's the mechanism. British pension funds were running these leverage strategies, and when yields spiked, they got collateral calls. To meet the calls, they had to sell GILTs. Selling GILTs pushed yields higher. Yields triggered more collateral calls. That's a doom loop in a sovereign bond market, in a G7 country. The Bank of England had to step in and buy long-dated GILTs from September 28th to October 14th in the middle of a hiking cycle. And they insisted the whole time that it was, quote, temporary and targeted. And definitely not quantitative easing. Yeah. The fig leaf is the tell. And then Quasi Quarteg, the Chancellor, sacked October 14th. Liz Truss resigns October 20th. 44 days. Shortest premiership in British history. A bond market removed a head of government in under a month in October. Now, here's the live part. Britain is doing it again right now. 30-year guilts have hit 5.8% on August 24th, highest since the late 90s. And JP Morgan attributes the sell-off to, in their words, an energy-driven supply shock. That's my link one going into my link two, in print, this month in a real country. And notice who this keeps happening to. In 1956, Britain finds out its reserves under constraint. 2022, Britain find out its pension plumbing is the constraint. Twice in 70 years, the same country got taught the same lesson. You don't find out what you actually depend on until you need it. That's the whole reason I do episodes like this one. Korea. Quick one, but I think this is the most single underreported thing in the global markets right now. South Korea's stock market went up 116% this year, to a record on June 22nd, then it fell almost 40%. Peak to trough in about five weeks. Circuit breakers tripped. Their version of the VIX hit 98%. Ours is sitting around 14.5%. And chips are about 44% of what Koreans sells the world. Same chips, same customers, same trade as ours. So one of those two markets already had its crash, and the other one is near a record high. I don't think that's because we're safer. I think it's because Korea is where the trade gets priced without a story attached to it. In 1931, the crack didn't start in New York. It started in Vienna, at a bank most Americans had never heard of, and it took five months to get here. If you want to know whether this rhyme is live, don't look at the SP. Look away from home. The AI loop. All right, the AI thing. And this is the direct sequel to the circular financing segment I did last time. So I'm not going to re-explain the structure. Nvidia reported Tuesday, 96.2 billion in revenue, 5.7% above their own guidance. 15th straight beat. Jin Chen Huang went on television and said it demanded super strong and accelerating. Wednesday, they paused their revenue sharing and credit support program for AI cloud companies eight weeks after they launched it. Now, those two facts do not sit comfortably together. That program was NVIDIA guaranteeing other people's debt, take or pay commitments, minimum revenue guarantees, agreeing to rent back unused capacity if demand fell short. It's what made the Neo Clouds financeable. NVIDIA was underwriting its own customers' ability to buy NVIDIA products. And I want to be fair here because there are innocent explanations. Accounting treatment, auditor pressure, the optics of the circular financing criticism. Maybe the thing was just rushed out, needed restructured. But whatever the reason, here's the effect. If demand really is accelerating the way he said, those guarantees are free money. You never have to pay out. You only pull back a backstop after you've looked hard at what you might have to cover. Withdrawing a guarantee is a company's own risk department voting with the balance sheet. And words are cheap. One more piece, and this one's underrated. The AI buildout is inflationary and not in a hand wavy way. PJM is the biggest power market in America. Their capacity prices went from about $29 a megawatt for 2024-25 to $329 for 2026-27. The most recent auction cleared at the regulatory price cap for the third straight year, and PGM's own independent market monitor attributed roughly $6.3 billion of the $16.4 billion in total capacity charges to data centers. That's not abstract. That's electricity, which is a core service showing up on your utility bill. So follow the loop. The AI boom bids up electricity and memory chips. Those feed the inflation number. That inflation is what forces the Fed to hike. And the hike kills the longest duration trade in the market, which is AI. It's eating itself, fueled by an energy crisis. Again, thank you, Donald Trump. The kitchen table. Let me bring this down to ground level because so far this has all been plumbing. I'm going to give you three things, and I'm going to say would instead of will, because none of this has happened yet. First, the pump in the shelf. AAA has the national average around four or and here's the stat. This is the first August ever where the national average stayed above $4 every single day of the month. Diesel matters more than people think because diesel moves freight and freight moves food. Second, everything with a variable stat. Prime is at 6.75 right now. If the Fed hikes, it goes to 7, and credit cards and home equity lines reprice on the next statement. Not eventually, next statement. The average HELOC balance is over $50,000, and most of those are variable. Third, and this is the counterintuitive one, in a normal recession, rates fall and that saves you. Your mortgage rate gets cheaper, you refinance, and the relief's val and that's the relief's value. In a plumbing crisis, it goes the other way. Yields rise while the economy weakens, and mortgage floods blow out on top of that. So the thing that's supposed to rescue you is the thing that breaks. We saw a version of that in March 2020. And all this would be landing at 4.1% unemployment. That's the cushion. That's the good scenario. Now, the Hoover tie here is the one that actually bothers me. In June of 1932, in the absolute depths of the Depression, Hubert Huber signed the largest peak sign tax increase in American history to that point. Top rate went from 25% to 63% to close the deficit during a depression. And not because he was cruel, because the orthodoxy of the time said a balanced budget was what restored confidence. The theory was about confidence. The bill landed on people. And one caution, because I got a couple of messages about this after the last episode. Most of the money in money market funds is in government funds. The fee rules apply to institutional prime and institutional tax exempt funds. Redemption rates were eliminated entirely in the 2023 reforms. I'm not telling you to move your cash. That's not what this is. No exit. So let's say I'm bright and something breaks. What happens? Last time I gave you the Fed's triple lock. Every tool makes something else works. I'm not going to rerun it, but I want to add the part I left out, which is Congress. The debt crossed $40 trillion on August 19th against an economy of about $32.5 trillion. That's roughly 123% gross. The debt ceiling is $41.1 trillion. It got raised by $5 trillion from $36.1. In the reconciliation bill signed July 4th of last year. So do the math. That's about $1 trillion of headroom, and we're already borrowing $1.9 trillion a year just to run the government. A TARP-sized emergency bill would consume most of the remaining borrowing authority in a single vote. And then they'd have to go raise the ceiling again in the middle of a crisis. Now I want to be honest about TARP because I think it was true last time. On September 29th, 2008, the House rejected the bailout. 205 to 228. Bipartisan revolt. Both parties hated it. That same afternoon, the Dow fell 777 points. Single biggest day drop in the history of time. Four days later, Congress reversed and passed it. So, Washington can act. That's the honest version. Washington just can't act in advance. It acts after the market has already inflicted the damage. My claim isn't that nothing can be done. My claim is about the price of the delay. And one precision point, because I got this wrong in my own head until recently, the Fed's emergency lending authority under Section 13.3 does require the Treasurer Secretary's permission. But the standing repo facility is a different thing. It's a board tool under Section 14, and it does not. So the emergency door needs Treasury's approval. The routine plumbing door doesn't, which is exactly why the fight over who's allowed through that second door matters so much. Last thing, Congress passed a funding bill on August 8th, 90 to 6. That funds the government through December 11th. They did it specifically to avoid a shutdown right before the midterms, which is completely rational. But look at what that means. There is no must-pass legislation between now and December 11th. No deadline, no vehicle. If something breaks in October, they'd be building the response from a st from a standing start. In the last three weeks of a campaign where control of both chambers is on the ballot. And here's the Hoover ending, which I think is about a lot. Hoover's answer to the banking crisis was the National Credit Corporation. In October of 1931, it was voluntary. The healthy banks were supposed to rescue the sick ones out of enlightened self-interest. It was slow, and it disappointed even Hoover. The Reconstruction Finance Corporation, the thing that actually worked, the thing with real federal money behind it, didn't arrive until January 22nd in 1932, three months later. And by then, 512 banks were already gone. Nobody was asleep. Hoover was working constantly. The tools arrived in the wrong order and the wrong sequence. The voluntary one first, the real one after the damage. That's not a story about stupidity. It's a story about what institutions can bring themselves to do and when. And here's the thing right now Donald Trump is in the perfect position, more so than any other executive in the history, to do whatever he wants, thanks to the Supreme Court and the unitary executive theory. And a Democratic Congress, or even one heavily close divided, is not going to hand him a blank check after he has spent almost two years just pissing them off. So you're going to have a Congress that's adversarial and with extra incentive to not pass anything, and a president with all the incentive to break all the norms that may or may not even be legal. And I'm not going to end without this part. There's a good objection to my whole thesis, and I want to give it to you properly. It goes like this The basis trade is hedged. It's long cash treasuries and short futures. So if it winds, they sell the bonds and buy the back the futures. And the net effect on yield is roughly zero. So why would anything blow out? Here's my answer, and it's about sequencing. Margin on the futures leg, margin on the futures leg gets called by a computer overnight on a volatility model. Repo haircuts on the cash side move slower. They're negotiated. So when the call comes, you sell the thing you can sell right now, which is the cash bond, and you take the hedge off later in pieces. Over the full unwind, it nets out. During the unwind, it absolutely does not. And during is the part that breaks things. And we have the receipt on that. In Britain in 22, they sold the cash guilt days before they unwound the derivatives. That's the documented sequence. Second thing, clearing houses have real anti-proseclivity protections right now, margin floors, volatility buffers, international standards that came in this year. Those are genuine and they smooth the cliff into a steep pole. But there's 1.8 trillion sitting on that slope. So I'm still worried. Five ways I'm wrong. The Fed widens who can borrow from the repo facility. Treasury buybacks actually work. There's a coordinated currency intervention that holds. Hormuz reopens, or the buffers just do their job. The full chain is maybe one in ten. The pieces are much liker than the whole, but judge me on the pieces, and I'm still betting at my chatbot and saying they're all gonna click. The scorecard. Okay, here is the full sequence of dates that I am looking at going forward. Okay, accountability time. I said last episode I'd give you the numbers, so here they are, and the definitions are locked right now, so I can't move the goalposts later. I'm going to read three on air and put all seven in the show notes because I don't want to read you a spreadsheet. One, the Fed hikes 25 basis points on September 16th. That is a prediction. And I want to be straight with you. When I first made this call, it was priced around 30%. It's 57% on Fed's fund futures as of this morning. This is the market's base case now. Not my brave prediction, and I'm not going to claim credit for it if it lands. Two, the Bank of Japan then hikes by 50 basis points on September 18th. Why does the Bank of Japan hike 50 basis points instead of 25? Because remember, the Bank of Japan is trying to defend their currency. To defend their currency, they have to narrow the interest rate gap between them and the Fed. If the Fed moves 25, then the Bank of Japan moving 25 does absolutely nothing to help the currency, which is the whole point of the Bank of Japan raising in the first place. They are structurally forced to raise at least 50 basis points in order for their rate increase to even do the thing they're trying to get it to do. Okay? This is just basic economics. But it's going to cause faster shifts and faster ripples than anyone else to predict. Three, and this is the one that actually matters. One single trading session between now and October 13th, where the 10-year treasury reeled goes up and stocks go down at the same time. That third one is the day the safe thing stops being safe. That's the tell. Everything else in this episode is an argument for why that day might come. Here's everything from the full show notes without me reading the spec sheet. Just to recap. Fed hikes 25 basis points September 16th. Bank of Japan hikes 50 basis points September 18th. The yin then rallies at 72 hours by September 21st. I'm putting Brent above 102 for five successive closes by September 30th because I don't think Hormos is getting reopened. Iran is existentially fighting for their existence as a state, just like Ukraine is currently fighting Russia. And Americans do not understand that. And the Trump administration does not understand that. They the af that's the same reason we lost Afghanistan, and we're not even willing to use the army this time. They do not care about whatever economic sanctions we throw at them, and they're not going to sign a peace treaty until after Donald Trump leaves office. It's the same thing they did to Jimmy Carter, and they're just going to do it to Trump. And Hormuz is going to stay shut. And it's going to be at 102 because by the end of September, because the SPR is going to run out, and the artificial supply the government's been pumping in will vanish. Number three, I've got CFTC leverage fund net shorts down 20% by October 2nd. And that's it. I'll grade all seven in October. But here's the problem. If any once these circuit breakers fall, like I said, the Fed hikes into a Bank of Japan hike, into yen rallying, into crude, into an AI stock mobile popping. I have the AI stock popping mid to late October, right before the election. I want to be careful landing this because I've spent an hour telling you that things could go wrong. And honestly, that's a genre I don't like very much. So let me be plain. I don't know for a fact that any of this happens. I'm just looking at the tea leaves and telling you everything is primed for it to happen. Two pieces of history tonight, two doing two different jobs. Suez is how the fire starts, Hoover is how the fire department makes it worse. In 1931, the Fed raised rates to defend its credibility, and October did the rest. If we choose credibility again, don't be surprised when the pipe ceased first. Warsh is determined to choose credibility. The Bank of Japan is determined to choose credibility. Same thing. And remember the part I mentioned earlier. In 1932, the Fed finally did the thing, April through August. They bought about a billion dollars of government bonds, real quantitative easing, 60 years before anybody had the phrase. And the research says it's worked, treasury yields came down. They stopped in August. Not because it failed, because they were worried about gold and what it looked like. The tools existed, they used it, it worked, and they put it down anyway, because using it conflicted with what the institution believed about itself. The tool was never the problem. The tool is never the problem. I told you last time to watch Warsh. Well, he spoke. He told us the underlying trend in inflation hasn't improved, and that if it doesn't, they have work to do. The market heard it and repriced in two hours. But he did not discuss the plumbing. Not the repo facility, not who can borrow from the Fed in the squeeze, not once. Nothing Bessent's been pressing him to talk about. In 1931, the Fed defended the currency and let the banking system go. And it wasn't malice. They were looking at the thing they thought was their job. On Friday, the chairman looked straight at inflation and straight past the pipes. Seven numbers with dates. If they don't print, I was wrong and I'll say so right here. All right. Thanks for listening and I'll see you next time.