In the first in a series of articles on structured credit valuations, R² Financial Technologies ceo Dan Rosen discusses the challenges and complexities of pricing structured credit securities, and the perils of doing so inadequately

Pricing models for structured credit securities have often been overly simplistic, with many institutions placing too much reliance on dealer quotes, on simple models based on credit ratings and on top-down views of collateral portfolios. As liquidity dried up in the markets, many practitioners were left in the dark, not able to reliably determine the value of their portfolios or to analyse their risk. In this series of articles, I point to the importance of a model risk framework, given the inevitable limitations of industry models, and discuss several lessons and best practices that are being re-learned as the global banking system copes with the legacy of the crisis.

The financial industry just experienced a (hopefully) 'once-in-a-lifetime' crisis, which nearly brought down the entire system. A recent report by the IMF estimates bank write-downs and loan provisions between 2007-2010 at US$2.3trn, with about two-thirds of these losses (or about US$1.5trn) realised by the end of 2009.

Originating first in the US subprime market before spreading around the globe, the credit crisis has highlighted many limitations of the industry's general valuation practices and our understanding and management of risk, particularly as they relate to structured credit portfolios. Market participants clearly misunderstood and underestimated the risks in many securities, especially with respect to systematic risk, default correlation, contagion and liquidity.

Modelling, valuation and risk measurement of structured credit products is challenging, given the complexity of the structures and underlying risks. Investors have in practice generally relied on periodic valuations by dealers or other third parties, or on simple models based on credit ratings and on top-down views of collateral portfolios.

This black-box approach has resulted in a lack of transparency in prices and limited risk capabilities (risk measures, stress testing, concentration risk, etc). During the housing and credit boom of the last decade, structured finance instruments generally performed well.

There was a widespread perception that their risks were small and contained. This view proved to be wrong...and costly.

Valuation and risk solutions for structured credit portfolios are computationally intensive and require important investments in methodology, systems, people and the integration of multiple data sources. Several difficulties in valuing and measuring their risk include:

• Complex risk profiles. Portfolios contain market risk (interest rates and spreads), credit risk and prepayment risk (as well as liquidity risk). These risks are intertwined and it is difficult to capture their interaction effectively. Correlations are very important and difficult to assess, and they are not widely used or standardised. Systematic concentrations have proven to be key drivers of losses, as well as possible contagion (within a market, as well as across markets).
• Complex structures. The cashflow structures are complex and generally opaque for a standard investor. They are difficult to model and are computationally involved.
• Lack of reliable pricing data. Pricing, as well as fundamental credit data, comes from multiple data sources and is often incomplete or unreliable. The lack of liquidity in the market increases this problem.

What is the value of a security and how do we know?
When dealing with complex structures and markets with limited liquidity, it is important to understand the meaning and use of a 'price'. It is vital to acknowledge the sometimes "heroic" assumptions in industry models and the limitation of the information that we can reasonably extract from the market.

When liquidity is thin, dealer quotes are unreliable and model parameters cannot be estimated based only on observed market prices. The crisis has made evident these limitations and has further highlighted the subjectivity of valuation models and their assumptions.

In practice, we must effectively incorporate into the valuation process fundamental credit information, historical data and expert judgment. In addition, it is vital to develop explicit model risk and stress testing approaches, which can help us understand better the behaviour of instruments and portfolios, together with their risks and the 'Knightean' uncertainties we are facing.

A quote by the famous dramatist and poet Oscar Wilde (1854‐1900) comes to mind:

"Nowadays people know the price of everything and the value of nothing."

If Wilde were alive today, his quote might well have read something like this:

"Before the crisis, people knew the price of everything and the value of nothing... now they know neither."

In upcoming articles, I will explore in more detail several lessons and best practices the market is re-learning as it emerges from the crisis, as well as the application of model risk and scenario analysis concepts to structured credit portfolios.