Handling Liquidity Risk in Investment Funds

By Vinodh Nagaiyan, Risk and Finance Solution Consultant, and Sindhurika Kalkunte, Product Strategy- Enterprise Risk Management, Oracle Financial Services

As of 2019, the worldwide mutual fund industry had $55 trillion in total assets, according to the Investment Company Fact Book.[1] This figure has seen a consistent increase in the past decade and is poised to grow further in the near future. The growth of this industry highlights the importance of risk management practices – particularly liquidity risk management.  

In a collective investment scheme, from the investor perspective, subscriptions and redemptions are smooth, mostly online. From the company perspective, however, redemptions are the outcome of a delicate balancing act. If the fund holds a large number of liquid securities to facilitate redemptions, then it misses out on returns. If it holds a low number of liquid securities, then it might not be able to honor redemptions on a timely basis. Every redemption request should be honored on time, to be both competitive in the market and comply with regulatory mandates. Liquidity risk management is thus a daily affair. 

Given the size of the industry, it is no surprise that regulatory bodies are already setting rules and regulations in this direction. The U.S. has passed the Liquidity Final Rule, which requires open-ended funds to classify their holdings into various categories. This same rule also asks companies to maintain a minimum pool of liquid assets and to set thresholds on holding illiquid assets.

From a technology perspective, a comprehensive liquidity management solution is one, which helps investment companies not only manage redemptions, but other outflows as well. It should aid in bringing inflows and outflows in a single picture, so that dynamically maintaining buffers and putting countermeasures in action, when required, will be possible.  The ideal solution also addresses regulatory concerns and statutory reporting. 

Such a solution, should support the following capabilities:

  • Asset liquidity classification and Buffer maintenance
    Most funds quantify each asset’s liquidity risk for better comprehensibility. One of the popular metrics is ‘Time to liquidate’ (TTL) measured in days. Some funds also use ‘Liquidation cost’ (LC) as an approach, which depends on the number of assets a manager is trying to liquidate. 

    Using TTL as a basis, the next step is to determine the criteria, which varies depending on the level of classification. Assets can be classified at multiple granularities, and some assets, such as treasury bills can be classified at a high level based on characteristics such as product type, maturity, issuer, guarantor, currency, etc. On the other hand, some other assets need to be further examined at the instrument level. An example of this is equities — each equity’s liquidity is different depending on characteristics, such as sector, bid-ask spreads, price changes, etc. 

    A buffer helps the fund to deal with unexpected liquidity shocks. The U.S. SEC Regulation directs funds to set aside a buffer for contingencies, especially for funds dealing with illiquid assets. This buffer is called ‘Highly liquid investment minimum’ (HLIM) and is analogous to the High-Quality Liquidity Assets (HQLA), which banks need to hold to withstand extreme liquidity shocks. The HLIM is the number of liquid assets that a fund needs to carry in order to hold out against liquidity shocks, mainly in the form of redemptions. The buffer should be such that, it could be easily liquidated with a minimal haircut in the event of a liquidity shock, and also such that, upon liquidation of this buffer, the liquidity profile of the assets for existing customers isn’t altered significantly. 
  • Liabilities forecasting

This aspect involves estimating all of the fund’s future liabilities. The major liability for most funds is redemptions, but other liabilities include margin calls, liabilities owed to banks, charges and fees owed to various associated entities, and operational expenses. Projecting net fund flows with reasonable accuracy is challenging for funds. However, with advanced machine learning (ML) techniques, it is possible to project fund flows, timings, and trends credibly with past data or peer funds’ data.

Projections start by identifying relevant variables, such as age, location, disposable income, risk tolerance, etc., and establishing a relation between the fund flows and the identified variables.
The relevant variables to be chosen are dependent on the fund and the targeted demographic. The relationship could be anything from simple regression to advanced ML-based algorithms. Once the relationship is established, a ML-model is then used to project fund flows throughout the considered time horizon.

  • Stress testing.

 Stress testing fund flows has become a necessity in most mutual funds. For riskier and less liquid portfolios, it is mandatory to simulate extreme yet plausible events which might affect the portfolio adversely. This simulation, and the corresponding countermeasures, test the strength of the investment scheme and identify potential weaknesses.

While stress tests have to be tailor-made for the portfolio under consideration, the basic steps are common. First, identify the risk factors affecting the fund, such as concentration risk, funding liquidity risk, redemption risk, etc. Second, create assumptions involved around hypothetical incidents that challenge the fund along the risk factors chosen. Third, build various scenarios that involve varying severities, magnitudes, timelines, and other parameters for the incidents. For the fourth step, execute the scenarios so that the impact on fund flows is observed. This is superimposed on the forecasted flows to arrive at projected cash flows which reflect the results of the scenarios. Lastly, once the hotspots are observed, countermeasures that suit the fund can be defined, employed, and modified according to the desired outcome.

Ultimately, the entire process of classification, forecasting, and stress testing is cyclical, meaning it should be repeated and refined often.

A liquidity solution having the above functional capabilities should also be:

  • Flexible, given that the industry is heterogeneous.
  • Built on a uniform data model and common architecture so that data from different sources is brought, processed, and reported together. 
  • Continuously compliant with multiple jurisdictions to help fund houses navigate the multitude of ever-changing regulations.

A comprehensive liquidity solution can aid investment companies to revamp the redemption and cash flow management process. By measuring, managing, and reporting liquidity risks, each fund can identify its structural strengths and reinforce them. Periodic reporting and data collection helps the regulator to foresee systemic shocks in the industry, which is a real possibility, given its enormous size and further growth prospects. The need of the hour is for companies and regulators to reinforce practices in assessment, monitoring, and eventually in the containment of liquidity risk. 

[1] Investment Company Institute 60th edition of the Investment Company Fact Book: A Review of Trends and Activities in the Investment Company Industry. Published in 2020. https://www.ici.org/system/files/attachments/pdf/2020_factbook.pdf