Where the Risk Lives Now
The demonstration ran this week
Three Turkish portfolio managers failed to meet redemption requests inside 4 days. Pusula Portfoy went first, then Tera Portfoy, which said two of its funds with a combined 366bn lira, about 7.5bn dollars, of assets had defaulted on redemption payments. Atlas Portfoy disclosed delays in its money market fund the following morning.
Tera is the interesting one. Its main fund returned more than 66,000 percent over 3 years by pouring billions of lira into a tight circle of related companies and amplifying the bets with borrowed cash. That is the whole architecture in one sentence: concentration plus leverage inside a structure that reports a value rather than discovering one.
The trigger was a rule change, not a market move. Turkey's capital markets regulator tightened limits on free-float ownership and affiliated-entity exposure at the end of August, and gave funds until year-end to shrink positions that breached the new caps. A forced seller was created by regulation, into stocks the same regulation acknowledges can be difficult to sell quickly.
The market response is the standard sequence. The Borsa Istanbul 100 fell as much as 7.7 percent on Wednesday and triggered a market-wide circuit breaker. Tera's chairman called it an extraordinary and unprecedented speculative attack. The regulator filed criminal complaints and imposed 2yr trading bans on named managers. The Financial Stability Committee convened and said the problems were temporary and manageable.
Read the mechanism and not the country
It is tempting to file this under emerging-market idiosyncrasy. The more useful reading is that Turkey is small, retail-funded and locally supervised, which means the same mechanism that operates quietly elsewhere was forced to produce a visible event.
Strip the names out and the anatomy is four moving parts. A position size the market cannot absorb. A valuation that is asserted by the manager rather than transacted. A funding leg that is borrowed against that asserted valuation. And a liability side that promises access to money the assets do not contain.
Every one of those four is a description of the private-mark complex at a scale Turkey cannot approach. The difference is that in the listed and retail Turkish version the redemption request arrives and the failure is observable within days. In the private version the same stress arrives as a revision.
The risk did not disappear, it moved
What the listed credit market is pricing right now is not stress. High yield OAS at 2.76 percent (as of 15 Sep) is 1 basis point wider than the prior session. Investment grade OAS has not moved in 2 sessions at 0.80 percent. HYG is down 0.56 over the week, JNK 0.65 and LQD 0.86; those are duration marks, not default marks.
The all-in cost of money has repriced violently against that. The 10yr closed at 5.01 percent (as of 16 Sep), while the 5yr sits at 4.86 and the 2yr at 4.74, up 7 basis points on the day. The price of default risk has not moved. The price of money has moved a great deal.
The gap is the point. Post-2010 regulation pushed risk out of the deposit-taking system and into funds and vehicles that are supervised differently, funded differently, and marked differently. That migration was deliberate and it worked, in the sense that the banking system is genuinely more resilient than it was. It also means the risk now sits where the reporting is quarterly and the mark is an opinion. The stress is being expressed in the equity risk premium and the long end. It has not yet reached the corporate borrower.
The fragile part is not the equity
The intuitive fear is that private equity marks come down. That is the less important half. The part with a price and a creditor is the leverage sitting underneath: subscription lines against uncalled capital, NAV facilities against the portfolio's stated value, and rated debt inside the private-credit complex that pension funds and insurers hold for yield.
Only the last of those has a daily mark and a margin mechanism. That is where the sequence starts. A mark is questioned, the NAV facility attached to it gets a collateral call, the holder has to post or sell, and the selling arrives in the same illiquid underlying that produced the mark. That is the same shape as 2020 and 2008: the fragility lives on the liability side against assets that cannot be sold at the printed price, and the failure is a funding failure before it is a credit failure.
The official sector's own research says this. The IMF's April 2026 Global Financial Stability Report describes private credit vulnerabilities in almost these terms: borrower weakness, multi-layer leverage, stale valuations and unclear linkages, with semi-liquid structures identified as a specific weak point and NAV facilities flagged for enabling leverage against marks. The Financial Stability Board's May 2026 report adds the sentence that matters most, which is that the market has not been tested through a severe downturn.
The official sector can see it
This is where the Don't Look Up analogy breaks down, and the correction is worth making because it changes what you would expect to happen next. Regulators are not refusing to look. They are looking, and concluding that pre-emptive action risks causing the repricing they are trying to avoid.
That is hostage logic rather than denial. Supervisory authority is national; the exposure is cross-border, channel-based and increasingly held by vehicles and investors that no single supervisor owns. So the instrument set shrinks to leaning against the risk, disclosure requirements, and capital treatment of the banks and insurers that lend to it, none of which unwinds the leverage itself.
Which sets up the actual next move. If the structure is too embedded to defuse quietly, the realistic official response to the first genuine funding stress is not to make whole the assets. It is to legitimise the gate: a facility, a rule, or a supervisory blessing that suspends redemption and lets the mark stay where it is. The loss is then socialised through the holder rather than the taxpayer, no price is printed, and the crisis is resolved by declining to discover it.
Three things that would make this more than a thesis
First, redemption pressure in the semi-liquid retail channel, where quarterly windows against illiquid assets are the structural mismatch and where the fund can only manage the flow by limiting it.
Second, and the cleanest single indicator, the gap between listed vehicles and the marks they report. If a listed private equity or private credit vehicle trades at a material discount to its stated net asset value, the market has begun to disagree with the manager's number, in public. That is the Turkish experiment with a ticker attached.
Third, the funding cost of the leverage leg rather than the assets. NAV facilities and subscription lines are floating-rate and referenced to the same short rate the Fed just raised. Every one of them repriced upward into a portfolio carried at a stale number, which compresses the excess spread that justifies the leverage in the first place. The structure does not need asset deterioration to become unprofitable. It only needs the borrowing cost to exceed the mark.
Why this is a credit piece and not an equity piece
The equity holder in a private structure can only lose what they put in, and in a closed-end fund with no redemption mechanism they cannot force a sale. The creditor can. The entire vulnerability is that leverage converts an unrealised loss into a funded, callable, contractual obligation.
Which is why the listed credit market's calm is not evidence of safety. It is evidence that the exposure has moved to where the price is set by negotiation and the stress is reported rather than printed. Turkey produced a visible default because its structures were retail and daily. Ours will produce a gate, a revision, and a filing.
The Fed Hiked Into a Supply Shock — why the all-in cost of money moved and the price of risk did not