Haircutting Non-cash Collateral

TL;DR

This paper introduces a parametric model for haircutting non-cash collateral, incorporating market risk, liquidity, and idiosyncratic factors to improve collateral management and regulatory capital calculations.

Contribution

It develops a novel parametric haircut model using a double-exponential jump-diffusion process, extending traditional value-at-risk approaches for better risk assessment.

Findings

Model captures market liquidity risk effectively

Enables sensitivity analysis and stress testing

Applicable to equities, securitization, and bonds

Abstract

Haircutting non-cash collateral has become a key element of the post-crisis reform of the shadow banking system and OTC derivatives markets. This article develops a parametric haircut model by expanding haircut definitions beyond the traditional value-at-risk measure and employing a double-exponential jump-diffusion model for collateral market risk. Haircuts are solved to target credit risk measurements, including probability of default, expected loss or unexpected loss criteria. Comparing to data-driven approach typically run on proxy data series, the model enables sensitivity analysis and stress test, captures market liquidity risk, allows idiosyncratic risk adjustments, and incorporates relevant market information. Computational results for main equities, securitization, and corporate bonds show potential for uses in collateral agreements, e.g. CSAs, and for regulatory capital calculations.