ESMA to release new liquidity stress testing guidance in early 2019

in Regulatory/Compliance, 19.10.2018

For the most part, asset managers deal with risks according to market best practices. But what about risks for which best practices haven’t yet been fully established? Liquidity risks are one such area, and consequently regulators have been focusing on providing more guidance.

To that end, in February 2018, following the work of the Financial Stability Board (FSB) and the International Organization of Securities Commission (IOSCO), the European Systemic Risk Board (ESRB) published its recommendations on liquidity and leverage risks in investment funds.

Investment funds, because they gather money from a wide variety of investors in order to put it into the market, represent a source of systemic risk. Open-ended funds are of particular interest to regulators, since they allow investors to frequently redeem money, potentially forcing asset managers to sell assets on a significant scale in a falling market, triggering spirals in price. Another source of systemic risk is illiquidity, which can expose assets to the risk of unexpectedly high redemption demands. Since market conditions can change very quickly, asset managers may end up having to rely on levels of liquidity less resilient than expected.

Liquidity risk management, for all these reasons, is thus immensely important.

To enhance the liquidity risk management framework, the European Securities and Markets Authority (ESMA), building on IOSCO’s and ESRB’s recommendations, plans to release guidance on investment fund stress-testing. Its first draft is expected by June 2019, and will probably define stricter rules for such tests.

Why does liquidity risk need further guidance?

An ESRB survey revealed that such tests, although conducted regularly by most EU fund managers, differ widely in complexity, frequency, and type of scenario used. For instance, many fund managers determine the frequency of stress tests based on an arbitrary firm-wide policy, rather than taking the fund’s specificities into account. Additionally, the scenarios tested are often not representative enough to properly capture risks arising from market turbulence or redemption flow spikes.

ESMA has previously published guidelines on stress test scenarios, but the guidelines only apply to money market funds on which consultations are open. The guidance expected in June 2019, on liquidity stress testing at the investment fund level, should have a broader scope.

The target capability

Ultimately, asset managers will need to be capable of simulating both historical scenarios and, since future crises are likely to differ from past ones, hypothetical scenarios. With these scenarios they should be able to evaluate how both old and new crises would affect the NAV, liquidity buckets, and redemption requests.

For historical stress tests, time windows should be varied to prevent outcomes being overly dependent on one window. Hypothetical scenarios may need to try to anticipate specific negative events such as government defaults or severe downturns in currencies, in order to predict the likely consequences on the fund. Forward-looking scenarios may also require the creation of a dashboard displaying changed risk factors and their correlation matrix. Such models should be able to generate probabilistic scenarios based on implied volatilities. Both historical and hypothetical scenarios may be single- or multi-factor.

On the asset side, models assessing the liquidity risk should include bid-ask spreads, volumes, market turnover, time to maturity, and number of counterparties active in the secondary market. With these elements, the model can estimate the time to liquidate and the price at liquidation. Additional information useful to include may be collateral arrangements and margin calls.

Focusing on the liability side, in order to identify realistic stress tests, IOSCO has emphasised the need to be able to spot patterns in redemptions, taking into account key investor information that could improve liquidity risk management.

For instance, a responsible entity may consider whether publicity about the relatively poor performance compared to its peer group might lead to an increase in redemption requests, even if lack of data could be an issue. Fund managers should make reasonable efforts to analyse their investor base to better understand relevant characteristics like risk appetite and to try to predict circumstances in which investors may wish to redeem. Such efforts should include, at least, considering the marketing and distribution channel and analysing historical redemption patterns of different clusters of investors. However, doing so will be tricky due to little information, particularly for UCITS. Stress tests on redemptions should therefore take into account possible additional liquidity tools that the fund is allowed to activate during stressed market conditions and should be calibrated based on a stability analysis of the fund’s liability, which itself depends on the investor base (retail, institutional, etc.) In most cases, redemption scenarios should not be entirely based on historical scenarios, since the majority of investment funds have never experienced pressure on the liability side yet. Statistical tools, then, can be used to complement the backward-looking approach.

The ability of funds to meet redemption requests can be assessed by conducting, for instance, reverse liquidity stress tests and weekly liquidity stress tests. The former approach is used to determine the maximum size of outflows the fund can stand without substantially modifying the allocation, and is useful to provide insights into the fund’s behaviour. Testing it up to its failure point will help managers understand vulnerabilities so they can take appropriate precautions.

Need more info?

Please feel free to contact me directly or visit our webpage.

Next up on the KPMG Blog:

Leave a Reply

This blog is pre-moderated which means that all comments are reviewed by a moderator before they appear. KPMG reserves the right not to publish any comments made.