Valuation risk
From Wikipedia, the free encyclopedia
 |
This article or section has multiple issues. Please help improve the article or discuss these issues on the talk page.
|
Valuation Risk combines aspects of data management, financial engineering and modelling and uncertainties related to the changing conditions of financial markets.
Valuation Risks have a direct impact on internal and regulatory compliance, counterparty exposure and performance management. Recently, valuation risks have led to reputational risks; potentially impacting credit ratings, funding costs and the management structures of financial institutions.
Valuation risks concern each stage of the transaction processing and investment management chain. From front office, to back office, distribution, asset management, private wealth and advisory services; transparency, integrity and reliability of valuations have become critical. Moreover, issues associated with valuation risks go beyond the firm itself. With straight through processing and algorithmic trading, data and valuations must remain synchronized among the participants of the trade processing chain. The executing venue, prime brokers, custodian banks, fund administrators, transfer agents and audit now share files electronically and try to automate such processes, raising potential risks related to data management and valuations.
The need to provide transparency and ensure the integrity and consistency of the data, models and processes used to process and report calculations within valuations is at the forefront of any risk management strategy within any financial institution.
[edit] Background
In recent years, the financial industry grew in volume and diversity at a staggering rate. The market competitiveness and progress made in financial engineering led to highly creative and innovative strategies where new products and new structures were offered at very fast pace on the market. As most innovations are first proposed on Over-The-Counter (OTC) markets, they tend to rely on financial models, sometimes combining several models together. Financial models typically build on underlying assumptions and require calibration to a breadth of scenarios, business conditions and variations of the assumptions.
The shockwave which keeps affecting the credit and capital markets following the burst of the US sub-prime real estate market in late 2007 has tried most underlying assumptions and had sweeping effects on a number of models that would unlikely be calibrated for extreme market conditions, or tail value at risks. This leads to an emergency call for transparency and assessments of exposure from the financial institutions’ clients, shareholders and managers, echoed by the regulators. In this process, it appears that market exposure and credit exposure intricately mix into a single notion of valuation risk.
[edit] Managing Valuation Risk
Valuation risks result from data management issues such as: Accuracy, integrity and consistency of static data. Accuracy and timeliness of information such as corporate events, credit events, or news potentially impact them. Streaming data, such as prices, rates, volatilities are even more vulnerable as they also depend on IT infrastructure and tools therefore adding a notion of technical and connectivity risk.
Current undertakings observed among financial institutions involve the setup of centralised data management platforms, open to multiple sources of static and streaming data where all financial instruments traded or held can possibly be defined, documented, priced, historised and distributed across the enterprise. Such centralisation facilitates data cleansing, historising and auditing. It allows organisations to define and control pricing and valuation procedures as required for compliance. For OTC instruments, the platforms also involve the definition and storage of underlying information such as yield curves and credit curves, volatility surfaces, ratings and correlation matrices and probabilities of default.
In addition, an important aspect of managing valuation risk is associated with model risk. In search of transparency, market participants tend to adopt multiple model approaches and rely on consensus rather than science. In the absence of deep and liquid market transactions, and given the highly non-linear nature of some of the structured products, the mark-to-model process itself requires transparency. To achieve this, open pricing platforms are linked to the centralised data warehouse (as discussed) above. Those platforms are capable of using multiple models, scenarios, data sets with various distribution and dispersion models to price and re-price under ever changing assumptions.
The final aspect of managing valuation risks relate to the action(s) that will be taken within the firm as a result of the assessments of exposures and sensitivities reported. Firms are becoming increasingly dynamic and proactive at reporting their exposure, potential implications and verifying whether those match the defined risk policies. The management of tail risks will now also be reviewed, as the aftermath of the crisis has highlighted that allocating economic capital weighted by a very low probability of occurrence of an event amounted to considering a normal distribution of events or simply overlooking the tail risk.
