Today, Bloomberg announced its Liquidity Assessment Tool, or LQA, which gives institutional investors a quantitative approach to calculating liquidity risk consistently across asset classes.
It is the first liquidity estimation tool to combine Bloomberg's rich financial data and machine learning techniques to calculate the multitude of relevant factors influencing liquidity.
Bloomberg LQA provides risk managers, portfolio managers, traders, and compliance officers with a standard definition of liquidity and a consistent approach to measuring the expected cost of liquidation for a specific volume of securities, and a desired time horizon. It also provides a score designed to indicate security-level liquidity with respect to liquidation cost and its distributions across different volumes.
Emerging from the financial crisis, regulators across the globe continue to scrutinize the quality and comprehensiveness of existing risk assessment processes and propose new standards for liquidity management on buy-and sell-side institutions. Establishing a systematic approach to measuring and reporting liquidity risk demonstrates a firm's commitment to robust risk management practices, comprehensive pre-and-post trade analysis and efficient regulatory reporting that meet new industry standards.
“Assessing liquidity risk is an essential business process for both buy-side and sell-side institutions because they need to assess the cost of capital for any asset they want to hold in their portfolio or on their balance sheet,” said Ilaria Vigano, Head of the Regulatory and Accounting Products group at Bloomberg. “Bloomberg LQA provides a consistent data-driven approach to measuring liquidity that helps our clients make more informed investment decisions, as well as simplify their regulatory reporting and risk management processes.”
Bloomberg Professional service subscribers can access Bloomberg LQA data for more than 130,000 global government and corporate securities. The same data can also be provided for enterprise use, which allows clients to override Bloomberg's default inputs with their own assumptions so the model considers their unique perspective on the market.
While Bloomberg LQA covers government and corporate securities today, the methodology that underpins the tool can be applied consistently across asset classes, to assess liquidity risk at the portfolio-level.
Machine Learning Powers Bloomberg LQA's Data-Driven Approach
Data availability and quality is critical for quantitative models to provide insightful estimates. This is a challenge in less transparent markets, such as fixed income, where there is limited or less reliable data available.
To overcome this challenge, Bloomberg assembled a multidisciplinary team of experts to consider all of the complexities associated with liquidity risk analysis. Bloomberg LQA uses machine learning techniques, such as cluster analysis, to dynamically identify and leverage transaction data for comparable securities.
The application of machine learning enables Bloomberg LQA to transcend the inherent limitations of the widely used bid/ask approach to measure liquidity risk, which can overlook important variables. By leveraging a larger set of factors, Bloomberg LQA can better assess liquidity across instruments and across time.