You priced your call options to the cent. Did you ever price your protection?
Does it pay for itself?
Does your hedge actually pay for itself?
Most assets sold as “safe” are a slow leak. Bonds, gold, cash sleeves — they dampen drawdowns, but they quietly tax your compounding in every year the storm doesn’t come. The question is not whether a hedge protects. It is whether protection costs you geometric return on net.
What a bad hedge costs
What a bad hedge costs over time
The cost is not the premium line. It is the gap between arithmetic and geometric return — the compounding drag a “store of value” imposes while you wait.
In Spitznagel’s run, allocating to a store-of-value safe haven lowered a portfolio’s median CAGR from 9.5% to 9.1% (−0.4%). The hedge did its drawdown job and still made the system poorer over time. That is the default failure mode: insurance you can feel, paid for in compounding you never see. We catalogue this drag across the available tail-risk instruments so the failure rate of each is stated, not implied.
A hedge that raises compounding
A hedge that raises your compounding
The target state is not “less risk.” It is more exposure and a higher growth rate at once.
A correctly specified hedge lets you hold a larger position in your core asset, not a smaller one — because the tail is capped elsewhere. You stop trimming the engine to survive the crash. The protection earns its slot by lifting the compound rate it is supposed to drag. This is the mechanism the tail-risk hedging architecture is built to deliver.
How the safe-haven test works
How the safe-haven test works
A safe haven is not a thing. It is a payoff — defined by function, not by asset class. The test is binary and runs on two properties at once.
The two-property test
A hedge qualifies only if it (1) pays off in the storms that actually break your portfolio, and (2) raises long-run geometric return on net. Mitigate the wrong risk, or pay too much for the right one, and it fails the second test even when it passes the first.
The clean statement: a safe haven is cost-effective when its geometric effect exceeds its arithmetic cost — a positive net portfolio effect (Spitznagel, M. (2021), Safe Haven, Wiley). In his framing, the insurance-style payoff raised median CAGR from 9.5% to 10.0% (+0.5%) while averaging 0% yield — and let the portfolio hold 98% in equities versus 64–72% for the weaker havens.
Stated condition
This is a knife-edge. The same work shows the effect is allocation-sensitive — “a pinch of salt”: too large a hedge allocation reverses the gain. Sizing is the whole game, and most hedges are mis-sized. Our quantitative methods specify the allocation band where the net portfolio effect stays positive, and the wider research layer sets out the assumptions behind it.
Objection
“Isn’t all insurance just a drag on returns?”
In arithmetic, yes — the insurance leg averages a loss. The point is that arithmetic is the wrong ledger. To replicate that +0.5% CAGR with a fixed-yield store of value would demand an asset yielding more than 30% per year. The insurance payoff delivered the same lift at 0% average yield. The drag is real; the net effect is positive. That difference is the entire thesis — and the portfolio diagnostic measures it on your own book.
This is the philosophy and test layer behind The Fail-Safe Circuit. Before we route capital, we run your book through the two-property test.
3 fields · 48-hour document · no call, no sequence.
Frequently asked questions
What makes a hedge a true safe haven?
A true safe haven is a payoff defined by function, not an asset class. It must pass two properties at once: pay off in the storms that actually break your portfolio, and raise long-run geometric return on net. A bond or gold sleeve that dampens drawdowns but lowers compounding is a store of value, not a safe haven.
Does tail hedging lower or raise long-term returns?
It depends entirely on specification. A mis-sized store-of-value hedge lowered median CAGR from 9.5% to 9.1% in Spitznagel’s (2021) run. A correctly specified insurance-style payoff raised median CAGR from 9.5% to 10.0% — a positive net portfolio effect rather than a drag.
What is the cost of carry on a tail risk hedge?
Cost of carry is the compounding you forfeit while the storm has not arrived, not the premium line you pay. The relevant cost is the gap between arithmetic and geometric return. Spitznagel (2021) shows an insurance payoff achieving a +0.5% CAGR lift at 0% average yield, where a fixed-yield asset would need to yield over 30% per year to match it.
Why does geometric return matter more than arithmetic return?
Arithmetic return averages each year independently, but compounding is multiplicative, so a single deep drawdown drags the whole path. A hedge can average an arithmetic loss yet still raise the geometric rate by capping the tail. Judging insurance on the arithmetic ledger is the error; the geometric ledger is the one that compounds your capital.
How do I apply the two-property test to my own portfolio?
Run your book through both properties: confirm the hedge pays off in the specific crashes that threaten you, then confirm it raises geometric return on net at its current size. Sizing is allocation-sensitive — too large a hedge reverses the gain — so the test is binary and size-specific. The diagnostic does this in three fields and returns a 48-hour document.