Jonathan Ingles: Ben, tell me you had a less depressing week than I did — I've been thinking about one number for three days.
Ben Okonkwo: Hm, depending on the number, maybe I should be worried too. What is it?
Jonathan Ingles: One point two five trillion dollars in derivative exposure. Four point seven billion in equity. That's Long-Term Capital Management — 1998 — before it collapsed.
Ben Okonkwo: Oh, that number.
Jonathan Ingles: Right — and look, that's what today is about. Leverage. Why do institutions keep using a tool that, the moment it goes wrong, doesn't just hurt them — it destroys them completely?
Ben Okonkwo: So let me put the basic mechanic on the table, because I think this is where the intuition actually breaks down for most people. Two-times leverage — hundred dollars equity, hundred borrowed, two hundred in assets. Ten percent gain on the assets? You've doubled your equity return to twenty percent. But a ten percent loss hits you for twenty percent of your equity. The asset is still worth a hundred and eighty dollars. You've just lost a fifth of what you put in.
Jonathan Ingles: And LTCM wasn't at two times.
Ben Okonkwo: Twenty-five to one, probably thirty — over a hundred and twenty-five billion on the balance sheet, and the derivative exposure on top of that reaching over a trillion. So when Russia's August 1998 default triggered a global flight to safety, the liquidity they'd assumed would exist to exit positions just — it wasn't there. And that's the structural trap. The asymmetry isn't just that losses hurt more. It's that the equity base shrinks with each loss, so the same price move does more damage each time it repeats.
Jonathan Ingles: So the floor is total loss. The ceiling is — theoretically — unlimited.
Ben Okonkwo: That's the asymmetry that makes leverage permanent in finance. The upside keeps people coming back. And the question we're really trying to answer today is whether continuing to use it, knowing that, is a failure of rationality — or whether the system actually makes it the rational move.
Jonathan Ingles: But the ceiling being unlimited is what makes this feel rational when you're inside it. Let me make this concrete. You put twenty thousand down on a hundred-thousand-dollar house. Mortgage covers the other eighty. House goes up ten percent — you've made ten thousand on a twenty-thousand stake. You just doubled your money.
Ben Okonkwo: Right — but the part that doesn't fit is what happens on the other side.
Jonathan Ingles: House falls ten percent. You've lost your entire deposit. The house is still worth ninety thousand — it still exists — but you are wiped out. That's leverage. One sentence.
Ben Okonkwo: And that's actually the clean version. What people miss is what happens after the first loss. Say you don't lose everything immediately — you lose forty percent of your equity. Now your denominator has shrunk. The same size price move hits a smaller base. So a fifty percent loss requires a hundred percent gain to get back to flat — not fifty. The math stops being symmetric the moment you're down.
Jonathan Ingles: Wait — say that again. Fifty percent loss, you need a hundred percent to recover?
Ben Okonkwo: Yeah. If you have a hundred dollars and lose fifty, you have fifty left. To get back to a hundred from fifty, that's a hundred percent gain. The loss percentage and the recovery percentage aren't the same number. And at higher leverage ratios that gap becomes — I mean, it becomes almost structurally impossible to trade your way out.
Jonathan Ingles: So the trap is architectural. It's not bad luck at that point.
Ben Okonkwo: Now it gets worse. Because the lender doesn't wait for you to trade your way back. When your losses reduce equity below their required threshold — the margin requirement — they either demand more capital or they sell your assets for you. Forced liquidation. And they're selling into a market that's already falling, which pushes prices down further, which triggers the next margin call.
Jonathan Ingles: So the lender's act of protecting themselves is the thing that accelerates the collapse.
Ben Okonkwo: Exactly that. And LTCM — this is why that number matters — they had derivative exposure of roughly one point two five trillion dollars sitting on four point seven billion in equity. When Russia defaulted in August 1998 and the flight to safety drained liquidity from the positions they needed to exit, there was no orderly way out. The Federal Reserve Bank of New York had to broker a three point six five billion dollar private bailout involving fourteen major financial institutions just to prevent the forced liquidations from cascading into everything else.
Jonathan Ingles: Fourteen institutions. Because if LTCM sells, all fourteen of their counterparties take hits simultaneously.
Ben Okonkwo: And that's the piece the house analogy doesn't capture — the interconnection. Your mortgage lender doesn't care what your neighbor does. But at institutional scale, everyone holds correlated assets, everyone shares counterparties, so one forced seller is everyone's problem. The BIS calls this directly: an investor's debt capacity is increasing in that of other investors. Meaning the system built the fragility in while it was building the leverage up.
Jonathan Ingles: And the question that actually keeps me up — frankly — is whether the people designing this system knew that and built it anyway.
Ben Okonkwo: And knowing it — right, that's the thing. Because the BIS doesn't just document that it happened. The finding in Working Paper 972 is that debt capacity is *increasing in that of other investors* — meaning the leverage boom and the fragility are the same event, not two separate ones. The system looks solid precisely because everyone is levered up together.
Jonathan Ingles: So the moment everyone decides it's safe is the moment it's most dangerous.
Ben Okonkwo: Exactly. And let me make that concrete — imagine a fixed-income portfolio manager, it's a Thursday afternoon in August 1998, she's holding spread positions that have been profitable for two years. Correlated assets, same counterparties as six other funds doing the same trade. Russia defaults overnight. By Friday morning the flight-to-safety move has cratered her collateral values. She gets a margin call. She has to sell. But the six other funds got the same call, so they're all selling the same assets simultaneously —
Jonathan Ingles: Into a market with no buyers.
Ben Okonkwo: No buyers. Because the buyers are also leveraged, also getting margin calls. That's the Brunnermeier-Pedersen mechanism — forced sales depress prices, reduced collateral triggers further margin calls, further sales. The spiral is self-feeding. And LTCM proved it wasn't theoretical: a hundred and twenty-five billion in assets, and the Federal Reserve Bank of New York had to physically convene fourteen institutions to prevent the unwinding from becoming everyone's problem simultaneously.
Jonathan Ingles: But isn't the counterargument — look, couldn't she have just exited before the margin call? Seen the Russia risk, gotten out early?
Ben Okonkwo: That's liquidity risk, and it's the part that breaks the intuition completely. The assumption baked into leveraged strategies is that you can exit at something close to fair value. But in a stress event, the market that was absorbing your position on a normal Tuesday simply — it's gone. The price you can actually get is far below fundamental value, and you're on the clock because the lender is forcing the sale. You can't wait for the market to recover. That's not an option when your margin call is due now.
Jonathan Ingles: So 'exit early' only works if you exit before anyone else knows they need to.
Ben Okonkwo: And given correlated positions across institutions — yeah, everyone needs to exit at the same moment. The window is essentially zero.
Jonathan Ingles: And the BIS says 2008 ran this exact same script.
Ben Okonkwo: Identical. Subprime losses triggered margin calls, forced asset sales pushed prices further down, contagion spread through the interconnected leveraged institutions — the BIS documented the whole cascade. Same spiral, different asset class, decade later. Which is what makes the tail-risk framing misleading. LTCM, 2008 — these aren't once-in-a-generation accidents. The leverage structure *generates* the synchronization that turns any shock into a system-wide event.
Jonathan Ingles: Frankly that's the thing regulators cannot say out loud. Because saying it admits the system is permanently loaded.
Ben Okonkwo: And the part that comes later makes this considerably worse — because after the GFC, the regulatory response tightened constraints on banks, and leverage didn't disappear, it just moved. Where it moved, and what that means for the next spiral, is — well.
Jonathan Ingles: That is an unsatisfying place to pause.
Ben Okonkwo: Unsatisfying, yeah — but here's where it actually breaks open. Because the regulatory story after 2008 is not 'we fixed leverage.' It's that we fixed leverage in banks. And the BIS Working Paper 972 finding is that since the GFC, non-bank financial intermediaries — hedge funds, money market funds, pension funds, insurers — their leverage has grown significantly. So the system didn't deleverage. It redistributed.
Jonathan Ingles: And the NBFIs are outside the regulatory perimeter.
Ben Okonkwo: Structurally, yes. The post-GFC framework — higher bank capital requirements, liquidity rules — was built for banks. The FSB's 2023 report names this directly. Archegos in 2021, the September 2022 UK gilt dislocation — both happen in the NBFI space, both show the same forced-liquidation spiral, and neither had an established intervention framework ready.
Jonathan Ingles: Hold on. UK pension funds in 2022 — that's not hedge funds. That's retirees' money.
Ben Okonkwo: Exactly that. And the mechanism was — okay, so UK pension funds were using liability-driven investment strategies with leverage embedded through derivatives. When gilt yields spiked in September 2022, collateral values fell, margin calls went out across the funds simultaneously, and the Bank of England had to step in. Three-day window. Same spiral, different actors. The bank-focused regulatory framework had no hook into it.
Jonathan Ingles: So the question I actually want answered — is this regulatory failure, or is it the regulation working exactly as intended? Because if the BIS and FSB are documenting this in real time and nothing structurally changes, that starts to look less like oversight and more like a permission structure.
Ben Okonkwo: I think — and this is where I want to be careful about the evidence — the capital efficiency rationale is doing real work here. Any institution using leverage isn't just gambling. The same equity supports a larger asset base. That's a genuine, defensible efficiency gain. And the competitive pressure piece means... I mean, no single firm can unilaterally walk away from it. If your peers are running 2x and you run 1x, you're accepting lower returns on the same capital base. That's not irrational.
Jonathan Ingles: It's a collective-action problem dressed up as a risk management question.
Ben Okonkwo: Right — and that's actually why post-GFC tightening on banks just exported the problem. Each individual NBFI picking up that leverage is making a rational call given the competitive landscape. The aggregate outcome is a systemic vulnerability that no single actor created and no single actor can undo.
Jonathan Ingles: And the moral hazard piece — frankly, the Federal Reserve Bank of New York convening fourteen institutions to bail out LTCM in 1998 didn't reduce the incentive to lever up. It increased it. It told the next generation: at sufficient scale, the downside gets socialized.
Ben Okonkwo: And that's the precise mechanism. The bailout precedent makes leverage more rational for the next actor, not less. Which means the regulatory response didn't just displace the leverage — it actively made the case for using it again. The upside stayed private. The downside, when it arrived, got shared. NBFIs are now running that same calculation with less regulatory visibility than the banks ever had.
Jonathan Ingles: No established intervention framework. So who actually pays when the next synchronized margin call runs through a pension fund?
Ben Okonkwo: That's — yeah, that's the question the FSB report doesn't answer. It catalogs the disruptions. Archegos, the gilt market. It doesn't resolve who absorbs it next time. And the honest level of certainty here is: we don't know, because there's no equivalent of the bank deposit guarantee sitting behind a hedge fund.
Jonathan Ingles: So the system keeps generating the problem, keeps documenting the problem, and the distribution question — who takes the loss — stays permanently unresolved. That's not a bug in the oversight. That's the architecture.
Ben Okonkwo: And that's — I mean, that's the pattern. The 1998 bailout, 2008, Archegos, the gilt market. Four events, different decades, different asset classes, same spiral. At some point the word 'surprise' stops being available. These aren't tail events. They're the rhythm.
Jonathan Ingles: And yet each generation treats it as a surprise.
Ben Okonkwo: Because the incentive to treat it as a surprise is — yeah, it's perfectly rational. If you name it structural, you've named the distribution question. And nobody in the room wants to answer who takes the loss next time.
Jonathan Ingles: Three days into this week I could not stop thinking about one number. One point two five trillion on four point seven billion. And I think what actually bothered me is that the number wasn't a warning. It was a demonstration that the system would catch it.
Ben Okonkwo: The Federal Reserve Bank of New York made sure of it. Fourteen institutions. And in doing that, made the number rational to run again.
Jonathan Ingles: Upside private, downside socialized. That's not a flaw they haven't gotten around to fixing.
Ben Okonkwo: No. And the honest level of certainty I can offer is: we don't know who pays next time, and nobody is designing the answer. That's probably a decent place to stop — genuinely appreciated the push on this one.
