Understanding credit risk in a chaotic financial environment
New research offers a model for credit risk that adapts recovery rates to market shocks and economic regimes, improving risk analysis and pricing accuracy
The collapse of Lehman Brothers in 2008 sent shockwaves through the global financial system, highlighting a critical weakness in how institutions understood and modelled credit risk. Investors who thought they understood their exposure discovered that recovery rates – how much could be salvaged from defaulted assets – were far lower than expected. This painful lesson demonstrated that traditional credit risk models failed to account for the dramatic impact of market shocks and economic regimes on recovery rates. Financial institutions today still grapple with how to properly model these complex relationships, especially as new economic uncertainties emerge due to climate change (leading to both physical and transition risks) and technological advancements (most notably, the widespread use of AI).
Recent research addresses this persistent challenge head-on. Published in the Journal of Banking and Finance, the research presents an integrated approach to understanding and pricing credit risk that accounts for the unpredictable nature of recovery rates during market chaos.
Co-authored by Professor Qihe Tang in the School of Risk and Actuarial Studies at UNSW Business School together with Dr Haibo Liu in the Department of Statistics and Department of Mathematics at Purdue University, their research offers a framework that moves beyond traditional assumptions about recovery rates for financial professionals tasked with assessing creditworthiness in an increasingly volatile world. Rather than treating recovery as a fixed percentage, the research acknowledges that post-default outcomes depend heavily on market conditions and economic regimes.
Why recovery risk matters more than we thought
The 2008 global financial crisis exposed major gaps in how banks modelled recovery rates. During the crisis, recovery rates for corporate debt showed unprecedented volatility, leaving many risk models outdated.
Industry conditions when a default occurs strongly influence how much creditors recover. During industry-wide distress, fire sales significantly depress the recovery values of defaulted assets – a pattern observed consistently across different economic cycles.
Traditional models often assume fixed recovery rates – for example, a 40% recovery for all senior unsecured bonds – but this approach ignores how market shocks can drastically alter recovery prospects overnight.
“The determination of recoveries to creditors during bankruptcy proceedings is always a complex process which no tractable, parsimonious model can fully capture,” said Prof. Tang.
How market shocks change the credit risk landscape
The Basel III framework highlighted the banking sector’s need to withstand shocks arising from financial and economic stress. This regulatory focus recognises that unexpected events can dramatically affect markets and rapidly shift both default probabilities and recovery rates.
Learn more: How good corporate governance reduces equity volatility
In their paper, Modelling and pricing credit risk with a focus on recovery risk, the researchers examine how these sudden shocks affect key risk factors, including default intensity, interest rates, and reference rates. Their model is flexible enough to describe the extent to which shocks can dramatically reshape risk patterns for financial institutions.
Market shocks affect not just the likelihood of default but the nature of the entire recovery process. When multiple companies default simultaneously during a crisis, legal proceedings are prolonged and asset liquidations happen in depressed markets.
“In credit risk modelling, it is crucial to incorporate shock risk, which is in the spirit of various contemporary regulatory frameworks in the banking, financial services, and insurance industries,” the researchers noted.
A two-part approach to recovery modelling
Prof. Tang and Dr Liu model recover payments using a hybrid structure with two components. One part depends on what has happened in the market leading up to default. The other part reflects unpredictable factors at the time of default, which vary depending on the economic regime (e.g., expansion, contraction, or crisis).
This approach recognises that market dynamics make recoveries highly unpredictable. As the researchers explained: “recoveries are often unpredictable due to various factors, including lengthy and costly legal processes (such as bankruptcy and liquidation), volatile market dynamics (that cause collateral values to fluctuate), market illiquidity (that can hinder the sale of distressed assets), and information asymmetry between borrowers and lenders (which adds an additional layer of uncertainty to recoveries).”
The model is probabilistic. The pricing framework allows for risk premiums on all relevant underlying risks and effectively captures how markets process information about default risk. While the mathematics appears to be complex, the idea is straightforward: recoveries should adjust with changing market conditions, and pricing should reflect that.
Breaking down risk into five key components
The model tackles five types of risk that typically get tangled together in credit instruments:
- Shock risk: sudden market-wide events
- Regime-shift risk: transitions between economic states
- Diffusion risk: normal day-to-day market fluctuations
- Event risk: the surprise when default actually occurs
- Unpredictability risk: the inherent uncertainty in recoveries
The key idea is to separate these types of risk within the modelling framework. The model helps risk managers identify the specific factors driving their credit exposures, enabling more targeted hedging strategies and improved portfolio diversification.
The approach uses standard techniques from pricing theory to capture risk premiums for each component, making it compatible with existing risk management frameworks while adding new insights about recovery dynamics.
Practical applications for risk professionals
The research tests the model using three common recoveries in the industry: recovery of face value, recovery of treasury value, and recovery of market value. For each approach, the team develops pricing formulas that convert defaultable securities to their default-free equivalents.
Subscribe to BusinessThink for the latest research, analysis and insights from UNSW Business School
These formulas help price various credit instruments, including corporate bonds, floating rate notes, credit default swaps, and total return swaps. Risk managers can use these tools to improve pricing accuracy and understand embedded risk premiums.
For credit analysts, the key takeaway is that recovery assumptions should evolve with market conditions. Static recovery rates can lead to significant mispricing, especially during stress periods.
Making this work in your organisation
For credit risk teams, this research highlights the importance of incorporating shocks and economic regimes into models of defaults and recovery rates. These risk factors may simultaneously increase default risk and decrease recovery rates, creating compounding effects that traditional models miss.
Trading desks can leverage these insights to price distressed debt and credit derivatives more accurately. By understanding which risk factors drive pricing across different market phases, traders can identify relative value opportunities and avoid overpaying for credit risk.
Learn more: Foreign banks and credit booms: Hidden harbingers of financial crises?
Risk officers can employ the framework to improve stress testing. This helps model how extreme events might cascade, providing a more comprehensive view of potential losses during crises.
Portfolio managers benefit from understanding how recovery patterns vary across industries and economic regimes. This can inform sector rotation strategies and credit quality decisions as the economic cycle evolves.
The research ultimately helps financial professionals move beyond traditional credit risk models toward frameworks that capture the complex reality of defaults and recoveries. In today’s uncertain economic environment, such a nuanced approach to credit risk can provide a meaningful competitive advantage.