David Sterling: Megan, quick question before we get going — have you ever looked at a number and thought, that can't be right, and then confirmed it was right and felt worse?
Megan Skiendel: Regularly. What's the number?
David Sterling: 940.
Megan Skiendel: Nine-forty what?
David Sterling: The leverage ratio at Long-Term Capital Management. $5 billion in capital sitting underneath $4.7 trillion in notional derivatives exposure. And — I mean, to make that number land — a leverage ratio at 10:1 means every single percent the underlying asset moves, your equity moves ten percent. Multiply that logic out to 940 and you understand why one margin call they couldn't meet required the Federal Reserve to coordinate a $3.6 billion rescue.
Megan Skiendel: One call. Not a series of bad bets — one call.
David Sterling: One. Because the unwind would have hit every major bank's collateral at the same moment. The systemic exposure was that concentrated.
Megan Skiendel: And look — what that tells me isn't that LTCM was uniquely reckless. It's that leverage at that scale rewrites what the word 'prudence' means. The symmetric amplification cuts both ways with equal force and nobody in the room wanted to name that out loud.
David Sterling: Which is the thesis — leverage isn't just a quantity problem, it's a structural one. The Financial Stability Board has spent years building out exactly how unmanaged leverage transmits shocks across the whole system, not just the institution that originates the exposure.
Megan Skiendel: So what we're actually trying to figure out today — is whether any of the tools we've built since 1998 actually change that structural fact, or whether we've just gotten better at describing the problem.
David Sterling: Well, that question only makes sense if you first understand what the tool actually does — and I'm not sure most people do, even now.
Megan Skiendel: Walk me through it. Plainly.
David Sterling: Okay — you buy a house worth a hundred thousand dollars. You put down ten thousand, borrow ninety. The house goes up five percent — you've just doubled your money. Your ten thousand is now twenty.
Megan Skiendel: And it goes down five percent.
David Sterling: Your equity is gone. Entirely. That's not a risk of leverage — that is leverage. The math is identical going up and going down. Same equation, reversed.
Megan Skiendel: Symmetric amplification. It's not a side effect — it's the definition. And honestly, the part that should stop people is that nobody hides this. The arithmetic is completely transparent. So the real question is why 1929 happened. Retail investors buying stocks on ten percent margin, which is your house analogy exactly, and when prices fell those margin calls hit and people were forced to sell. Mass liquidations. The crash deepened *because* of the forced selling, not despite it.
David Sterling: And that forcing mechanism — that's the margin call. That's the part people miss.
Megan Skiendel: Right — but it's not just that losses amplify, it's the *timing*. The margin call forces you to sell at exactly the moment when prices are already falling. You're not choosing to sell. You're being sold.
David Sterling: Mechanically forced liquidation at the worst possible price.
Megan Skiendel: And everyone else who borrowed to buy the same assets gets hit at the same time. Which is — I mean, that's not a 1929 problem. That's a 2021 problem. Archegos, total-return swaps, multiple prime brokers simultaneously unwinding concentrated positions. Over ten billion in losses across Credit Suisse, Nomura, others. Same mechanism, different instrument.
David Sterling: The math was known in 1929. It was known in 1998. It was known in 2021. Frankly — that's the puzzle. Symmetric amplification has been fully documented for nearly a century and the surprise keeps recurring. What does that tell you?
Megan Skiendel: It tells me the problem was never the math. The math is the easy part. The hard part is sitting in a room where everyone's profitable and being the person who says — actually, the equation runs both ways.
David Sterling: And that's the thing nobody teaches — the surprise doesn't come from ignorance of the equation. It comes from the fact that the feedback loop builds the problem automatically. Nobody has to make a reckless decision. The system does it for them.
Megan Skiendel: Wait — say that again. The system does it?
David Sterling: Mechanically. Rising collateral values permit institutions to borrow more. They borrow more, they buy more. Purchases push prices higher. Now collateral is worth even more. So they can borrow more again. Nobody pulled a lever — the leverage expanded on its own. That's procyclicality.
Megan Skiendel: And this is — I mean, it's almost elegant in how terrible it is. Because the leverage is largest precisely at the moment when the asset is most overvalued. You're maximally exposed at the peak.
David Sterling: Largest at the peak, most violent at the trough. Because when sentiment reverses — every single step runs backward simultaneously. Margin calls force selling. Selling drops prices. Dropped prices trigger more margin calls. You can't stagger that unwind. It's concentrated deleveraging, all at once.
Megan Skiendel: Which is exactly 2007 and 2008. CDOs, repo financing, off-balance-sheet vehicles — all of it connected through the interbank liability network. One node fails and the loss travels to every counterparty simultaneously.
David Sterling: And that network is what makes it systemic rather than just — bad. The interbank liability web means losses don't stay where they originate. They propagate. The U.S. banking system in 2008 wasn't hit by one shock — it was hit by the same shock arriving through twelve different counterparty channels at once.
Megan Skiendel: And the unwind is most violent exactly when markets are least liquid. That's the trap. The fire-sale cascade floods the market with supply at the moment when there are no buyers.
David Sterling: Right — and that part's been documented since at least the 1930s. The Financial Stability Board knows this. The ECB has examined it. The procyclicality bind — leverage tightening margin requirements at exactly the moment institutions need liquidity most — that's not a new finding.
Megan Skiendel: Which is the question that honestly should bother everyone. If we've known about this structural trap since the 1930s, why hasn't it been designed out of the system?
David Sterling: I think — and this is the part I can't fully close — the answer is that you can't regulate a phase transition. You can only prepare for when it happens. The system is self-reinforcing until it isn't, and the math reverses symmetrically. You can't outlaw the reversal.
Megan Skiendel: Although — and this is the part that gets worse before we're done — the deeper problem might not even be the phase transition itself. It's that the regulation is measuring the wrong thing entirely. And what that means for derivatives, off-balance-sheet structures, the whole question of what's actually inside the regulatory perimeter — that's coming.
David Sterling: The perimeter question. That's the one.
Megan Skiendel: And the perimeter problem is — look, you'd think after LTCM, after 2008, after Basel III, we'd at least be measuring leverage correctly. But Archegos happened in 2021. Thirteen years after the crisis. With all the frameworks in place.
David Sterling: What instrument?
Megan Skiendel: Total-return swaps. Which is — honestly, the structure is almost elegant in how much it hides. You receive the full economic gain of owning an asset. The dividends, the price appreciation, all of it. Without the asset ever appearing on your balance sheet. No ownership. No disclosure. No leverage threshold triggered.
David Sterling: So the exposure is real. The accounting is invisible.
Megan Skiendel: And here's what ESMA actually documented in their Archegos case study — a single family office accumulated systemic-scale leveraged exposure across multiple prime brokers. Credit Suisse, Nomura, others. None of them saw the full picture. Each desk thought they were the primary relationship.
David Sterling: Wait — none of them knew the aggregate number?
Megan Skiendel: Bill Hwang may have been the only person who understood his own aggregate leverage ratio across all brokers. The prime broker desks didn't ask. Or — actually, I think the more accurate version is they didn't have a mechanism to ask, because the swaps weren't reportable under existing thresholds. So when the positions moved against him and everyone unwound simultaneously, losses hit ten billion across those banks. Same fire-sale cascade. Different instrument than 2008.
David Sterling: That's the structural gap. Not that the regulation was weak — it's that the exposure migrated into an instrument the rules weren't written to see.
Megan Skiendel: Which is exactly the LTCM problem dressed differently. LTCM's notional derivative positions were orders of magnitude beyond what any balance-sheet ratio captured, and the Fed still had to coordinate a $3.6 billion rescue because the unwind cascaded everywhere.
David Sterling: The FSB and the ECB — their argument is that properly calibrated capital requirements, the Basel III leverage pillars, are the durable fix. Sufficient loss-absorbing equity reduces the probability the cascade even starts.
Megan Skiendel: Right — but that only works if the leverage is inside the perimeter being measured. Total-return swaps aren't. Off-balance-sheet vehicles in 2008 weren't. The capital pillar is sitting next to the exposure, not on top of it.
David Sterling: So the ESMA position — that regulatory capital metrics systematically undercount real leverage — that's not a critique of the framework's logic. It's a critique of its jurisdiction.
Megan Skiendel: And the question that doesn't resolve cleanly is — even if you extend the perimeter, whoever's building the next instrument is building it to be outside whatever perimeter you draw. That's not paranoia. That's the incentive.
David Sterling: The perimeter moves. The exposure migrates. And the next vehicle might be something regulators can't even classify yet. That's — frankly, that's not a solvable problem. That's the permanent condition.
Megan Skiendel: And that's the part I keep sitting with. The pattern runs 1929, 1998, 2008, 2021 — margin loans on stocks, concentrated derivatives, CDO chains in repo, total-return swaps. Different decade, different instrument, and yet — expansion on rising collateral, correlated positioning, sentiment flips, margin calls, simultaneous forced selling, illiquidity cascade. The FSB post-mortems from 2010 read like the regulatory post-mortems from 1929. Structurally identical.
David Sterling: Identical. Which means the next one follows the same script. The question I genuinely can't answer is where it originates — retail margin accounts, a non-bank lender, some synthetic vehicle nobody's named yet, an emerging-market structure the current regulatory perimeter can't see until the cascade is already moving.
Megan Skiendel: Honestly — probably all four at once.
David Sterling: Which brings me to the question I actually wanted to put to you at the start. Which of these collapses — 1929, LTCM, 2008, Archegos — happened because the decision-makers didn't see it coming? And which happened because someone in the room saw it perfectly well and took the leverage anyway because it was profitable that quarter?
Megan Skiendel: Both. Sometimes the same room, same quarter. The LTCM partners understood the 940:1 number — they built it. And they believed the model enough to hold it. Archegos — look, Bill Hwang understood his aggregate exposure better than any prime broker desk that was servicing him. The banks didn't ask. They were profitable on the relationship. So I don't think it's ignorance versus recklessness. I think it's that the incentive to stay in the trade is stronger than the incentive to exit it, right up until it isn't.
David Sterling: Which is — I mean, that's where we started, isn't it. 940. A number that can't be right, that turns out to be exactly right. The math was always visible. The leverage ratio was always the same equation. What changes is whether anyone in the room wants to be the one who says the equation runs both ways.
Megan Skiendel: And nobody does. Not when it's profitable. That's the part I'm still sitting with.
