Hedging Large Equity Blocks with Portfolio Risk Models

Summary

The document asks how a dealer might hedge a large single-stock position acquired in a client block trade. The dealer is described as seeking to reduce the risk carried on its balance sheet, potentially by shorting other spot equities. The author wonders whether portfolio optimization methods, such as minimum-variance optimization or hierarchical risk parity, could determine the hedge.

No answer, model specification, data, or empirical evidence is included, so the document does not teach a settled hedging procedure. Its useful focus is the distinction between neutralizing a position’s market exposure and choosing a broader portfolio hedge: a practical approach would need to define the risk target and the assets available for hedging. The questions about models and resources remain open, and the suggested optimization methods are proposals rather than documented techniques used by dealers.

Key ideas

• A dealer taking a client block trade may seek to reduce the resulting equity exposure.
• The author is interested in hedging large positions using spot equities.
• Minimum-variance optimization and hierarchical risk parity are raised as possible approaches, not confirmed practices.
• The document provides no implementation details, evidence, or answer about dealer hedging models.

Tags

# Hedging large single asset positions


# Hedging large single asset positions












I recently came across an article that described how big market participants like GS, JPM, etc. take off large equity positions (block trades) of their clients, and putting that risk on their own balance sheet. Then they use their prop models to compute what assets they have to short to essentially be market/delta neutral again. I am mainly interested in how they do it using spot equities.

Does anyone know what sort of models are used in this case? I was thinking they were using some sort of portfolio optimisation models such as Hierarichal Risk Parity or minimum variance optimisation. Any help, resources, code is highly appreciated!

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.