IGUAKO Capital is a work of institutional fiction by the Iguako Institute for Applied Unreality. No financial services are offered, no accounts exist and nothing here is an offer, a contract or advice.About this work
George Town · 36 cities · English Domiciles Documents Iguako Network Site Index
  • Insights
  • 2024
  • Hedging what the origination book actually owns

Insights · Quantitative Markets

Hedging what the origination book actually owns

A valuation is not a hedgeable exposure. A signed commitment and a signed sale are. This note sets out the three hedges the firm permits, the test that stops a hedge being placed, and the two numbers the Investment Committee reads side by side every quarter.

Date
2024-08-20
Author
Anneke Thorsby-Adeleke
Head of Quantitative Research, Sydney
Division
Quantitative Strategies
Sector
Quantitative Markets
Reading time
5 minutes

Key points

Hedge signed commitments and signed proceeds, and leave the carrying value of an unlisted position open.

A hedge costing more than a quarter of the accepted loss is not placed, and the exposure is reported instead.

Proxy hedges are retired in the month their correlation fails, without waiting for a committee.

The firm's market exposures are not the exposures its balance sheet carries. Thirty-nine positions and twelve credit facilities sit in eight currencies, most of them unlisted, most of them illiquid, and none of them marked by a screen. The Quantitative Strategies division is asked to hedge what can be hedged and to price what cannot. The first discipline is saying which is which. The carrying value of an unlisted position is not a hedgeable exposure. A committed drawdown in six weeks, and a sale signed with a nine-month long-stop date, are.

Policy permits three hedges and no others. Committed but undrawn capital is hedged to the drawdown date once the commitment is signed. Proceeds are hedged once a sale is signed and the long-stop date is known. Funding cost on floating-rate facilities is hedged where the facility exceeds US$15 million and the tenor exceeds three years. Everything else is left open and reported. The rule that does most of the work is the last one. An open exposure written into a report is managed. An exposure hedged by habit is a cost nobody defends.

Several currencies the firm originates in have no depth beyond twelve months. Quotes exist, and the spread paid on a nine-month forward in one of them ran between 180 and 340 basis points annualised during 2023. The division measures that spread against the loss the firm has already said it would accept on the underlying position. Where the hedge would cost more than a quarter of the accepted loss, the hedge is not placed. The exposure goes to the Risk & Valuation Committee instead, with the arithmetic attached.

Proxy hedging is permitted and tightly constrained. A currency with no usable forward may be hedged through a correlated liquid currency, but only where the rolling three-year correlation exceeds 0.7, and only to half the notional. The division re-estimates those correlations monthly and retires a proxy in the month the correlation falls below the threshold, without waiting for a committee. Two proxies were retired in 2023 for that reason. A proxy that has stopped working is worse than no hedge, because it is carried in the reports as protection.

Execution runs through two prime relationships, Barangaroo Prime Services in Sydney and Aldgate Prime Services in London, under collateral agreements the Chief Risk Officer reviews once a year. Notional outstanding at 30 June 2024 was US$164 million. The proprietary balance sheet stood at US$498 million at the end of that year. Margin calls are funded from a dedicated liquidity buffer rather than from operating cash, because a hedge that forces the sale of the asset it protects has inverted its own purpose.

The programme cost 41 basis points of hedged notional in 2023, including spreads, roll cost and collateral drag. That figure goes to the Investment Committee every quarter next to the losses avoided, which were US$7.4 million on two currency moves and nothing at all on the other six currencies. A hedging programme that never appears to pay for itself is either well designed or unnecessary. The only way to tell is to keep both numbers in the same table and read them together.

Published 2024-08-20 by the Quantitative Strategies division. Research is prepared for eligible counterparties and does not constitute advice.