Advancements in liquidity measurements as the battle for deposits intensifies

Advancements in liquidity measurements as the battle for deposits intensifies

By Greg Muenzen, Director, Novantas.

Greg, thank you for taking the time to speak with us today. Please can you provide our readers with a quick overview of your experience and what your current professional focus is at the moment?

I have been a management consultant to the financial services industry for more than nine years, focusing primarily on treasury and risk issues. I have worked with financial institutions ranging from small community banks to international money center banks across the U.S., as well as larger institutions in Canada, Australia, and Latin America.  My team specializes in analytically-focused engagements around liquidity, asset/liability management, interest-rate risk, stress testing and funds transfer pricing. A fundamental component of our work is the behavioral modeling of deposit and loan portfolios that require constant innovation in analytic and econometric approaches.

What are the benefits of attending a conference like Liquidity Risk Management USA and what can attendees expect to learn from your session?

The Liquidity Risk Management USA conference is a good opportunity to interact with and learn from a broad range of liquidity risk practitioners and other stakeholders from a range of financial institutions, regulators, and consultancies.  My session will feature observations around current retail and commercial deposit trends from Novantas’s proprietary data sources, such as our Comparative Deposit Analytics and Commercial Deposit Study. I will also discuss insights from liquidity risk-focused engagements over the past year that are increasingly drawing on machine-learning techniques.

What role does machine-learning play with liquidity measurements? Is there a potential for banks to do more?

We are living in the age of big data, and as market interest rates rise, customer behavior is evolving.  In that respect, machine-learning techniques can be useful for parsing large datasets to draw conclusions and automating the refresh and recalibration of those conclusions. For example, machine-learning algorithms can be used to make sense of account- and transaction-level data to understand which customer balance segments have higher or lower liquidity lives and how these behaviors are changing as rates rise. This is all  essential for understanding the profitability of deposits.  A streamlined approach is critical for exercises around commercial depositors who can interact with an institution through a variety of treasury-management services.

What are some of the best practices you recommend banks adopt and consider when using stressed liquidity analytics in management applications?

From my perspective, one of the most critical requirements is ensuring that the bank’s funds transfer pricing framework pushes the cost of holding liquidity to the businesses on an appropriately granular level. This will help to guide profitability measurement and margin trade-off decisions for pricing.  Stressed liquidity assumptions directly inform these costs, and we have seen the exercise to calibrate assumptions mature over time. It is a recognition that while LCR prescriptions are ultimately the binding constraint, an economic view developed in-house that produces more granular insights can better support pricing and profitability applications.

Looking to the future, what key challenges do you see for liquidity risk professionals?

One significant concern is whether banks are positioned to monitor customer behavior on a timely basis given the direction of market interest rates—particularly deposit balance flows between on- and off-balance sheet products. We are seeing what was long hypothesized in the flat-rate environment rate sensitivity is picking up across consumer, commercial, and wealth deposit segments, and liquidity models calibrated in the flat-rate scenario are not predicting customer behavior accurately. Another challenge is the need to set liquidity treatments for new products. For example, how should online deposit balances gathered from a new direct bank launch or commercial balances gathered from a new hybrid DDA product offering be valued in the absence of internal observations?

You may also be interested in…

Sign up for your free Risk Insights account: