Using data to find a solution
Stress testing and scenario analysis are part of any robust liquidity risk management program. Regulations in both the US and EU incorporate stress testing, posing an interesting challenge for participants looking to account for both historical liquidity events and potential simulations.
Given this background, let’s discuss these recent regulatory mandates in more detail - starting with the background on why global regulators care about this topic. Fundamentally, a lack of liquidity is at the epicenter of most financial crises (with the 2008 crisis no exception). It also represents significant systemic risk to the capital markets when liquidity dries up at the same time that everyone is heading towards the exits and trying to raise cash. Regulators across the globe have introduced regulations to modernize liquidity risk management impacting both the buy- and sell-side. For example, the SEC explicitly referenced the Third Avenue fund failure multiple times during its adoption of its liquidity risk management rule.
Regulators’ focus on liquidity is exemplified by both the US SEC’s adoption of Rule 22e-4 (where the final rule “requires a fund to take into account relevant ‘market, trading, and investment-specific considerations’ in classifying its portfolio investments’ liquidity”1) and ESMA’s adoption of ‘Guidelines on Liquidity Stress Testing in UCITS and AIFs’.
Both the US and EU versions for their respective funds industry incorporate stress testing into its requirements. While the US Rule 22e-4 is more prescriptive than the ESMA version, both are highly principles-based when it comes to stress testing. The SEC release states “the requirement to take into account relevant market, trading, and investment-specific considerations achieves this goal and is broad and flexible enough to be relevant for all investment strategies and fund risk profiles. We continue to believe that the proposed classification factors could help funds in valuating relevant market, trading, and investment-specific considerations, and thus we have included guidance on many of these areas.”2 Note that the factors are guidance, and not prescribed. While more principles-based than the SEC version, ESMA has some fairly specific requirements, including that Liquidity Stress Testing (LST) “should employ hypothetical and historical scenarios”. The rule goes on to state “historical scenarios for LST could include the global financial crisis 2008-2010 or the European debt crisis 2010-2012.”3
The challenge, is to replicate factors that impact liquidity assessment from a historical event at least a decade ago. In addition to factors stated in my initial blog (volume or reasonably anticipated trade size, price impact or significant value impact, and time) other inputs are more universal and not client-specific preferences. For example - bid-ask spreads, price volatility and market accessibility are variables that should be able to be stressed in a robust liquidity model. These variables are explicitly supported within the ESMA guidelines: “…in the estimation of the liquidation cost and time to liquidation under stressed conditions, which are typically characterised by higher volatility, lower liquidity (e.g. higher bid-ask spreads) and longer time to liquidate”4 for the first two parameters and that “it should not be assumed that the portfolio can be liquidated at the full average daily traded volume of an asset…”5 for the market accessibility parameter.
ICE Data Services’ Liquidity Indicators services has put much thought into how to best handle these situations, and we look forward to working with industry to raise the standard for liquidity stress testing from a workflow perspective.
1 Page 90 of the SEC adopting release
2 Page 103 of the SEC adopting release
3 Page 37 of the ESMA adopting release, section V.1.8
4 Page 39 of the ESMA adopting release, section v.1.11
5 Page 38 of the ESMA adopting release, section v.1.9