Multi-firm review of liquidity risk management at wholesale trading firms

Multi-firm reviews Published: 10/03/2025 Last updated: 11/03/2025

This paper summarises our observations from the multi-firm review of liquidity risk management at a range of wholesale trading (sell-side) firms, particularly brokers, in scope of the Investment Firms Prudential Regime (IFPR). It sets out the good and poor practices that we identified which similar firms can reference to strengthen their approach to liquidity risk management.

This review will be of specific interest to sell-side wholesale firms trading on their own behalf and/or offering their clients access to primary and secondary markets in securities, metals, energy, and commodities. This includes commodity clearing brokers, principal trading firms (PTF) trading in commodities, inter-dealer brokers, and contracts-for-differences firms.

1. Introduction

Over the past few years, there have been various stress events such as the COVID pandemic, the Russia/Ukraine war, the nickel price spike, energy price volatility, the ION outage, and the failures of Credit Suisse and Silicon Valley Bank. During these events, we engaged with firms that had experienced instantaneous and idiosyncratic (firm specific) liquidity shocks. These included: 

  • large cash outflows due to margin-calls:
    • initially from central clearing counterparties (CCP) and over-the-counter (OTC) counterparties who had implemented same day margining, and
    • significant intraday/inter-day utilisation of liquidity capacity (either liquid assets or liquidity facilities) during the interval between paying margin-calls to CCPs and receiving margin from clients. We have observed firms having to meet numerous CCP margin calls across consecutive business days, significantly draining their liquidity, before receiving the first day’s margin from their clients at the end of the second day
  • buy-ins of large open short settlement positions
  • instances of poor management of client relationships, including inadequate knowledge of clients’ business profiles. In some cases, revealed hitherto unknown concentrations of client risk: within firms (eg, not appreciating that two or more counterparties were in fact related) and across firms (not appreciating a counterparty’s total exposure). We have also seen firms (and their stress assumptions) underestimate concentration in markets that may have a limited group of buyer and seller types that may in practice be likely to react to events in a similar fashion
  • we have also seen the importance of stressing assumptions around liquidity, whether that be the depth available in markets to accommodate stressed sales without major price impact or, on the liability side, the speed with which customers or counterparties can withdraw money if terms allow them to do so

In our Dear CEO letters to sell-side firms – our strategy for supervising wholesale brokers (January 2023, and January 2025), our supervisory strategy for principal trading firms (August 2023), and wholesale banks portfolio analysis and strategy (September 2023) we said that: 

  • firms are failing to develop their own competence in liquidity risk management and continue to underestimate their exposure to intraday liquidity risks
  • operational or reputational risks can quickly result in broader concerns about a firm’s safety and soundness, and the clarity of a firm’s communication becomes even more important
  • risk management has the potential to dampen or amplify harm to market integrity in times of stress. Firms should ensure their stress assumptions are updated in the light of market events and are fit for the current environment. Stress testing should recognise that severe stresses will often affect the entire system – true in respect of market exposures, but also operations and reaction times, including in services provided by third parties
  • impacted firms described shocks as extreme even though previously observed volatility had reached similar or greater levels
  • PTFs who specialise in commodity market trading in particular may have been exposed to periods of sustained volatility, high prices leading to substantial margin calls and other higher costs, credit stresses and increased counterparty risks. While these firms have few, if any, customers, some of them have more complex interactions with markets and the real economy
  • many firms have put in place remediation programmes in response to these stress events. Better firms will have done this whether they were directly affected or not. We will look to senior management to evidence how these programmes have delivered better risk management and oversight across businesses and how they are comfortable that this is underpinned by a strong culture. We will also look to boards to evidence how they are ensuring that such improvements are lasting
  • we expect firms to proactively manage their outsourced arrangements
  • we expect firms to understand their dependence on third party providers and take steps to mitigate the potential impact on business continuity that the loss of service may have
  • good risk management protects profitability. 

Sell-side wholesale firms subject to the IFPR should, by definition, not be globally systemic. However, many of these firms are key participants in certain specific markets like commodities, metals, and energy. In some cases, the structures of these markets differ from equities, fixed income or derivative markets which have been subject to material reform since the global financial crisis. A disorderly failure of one or more of these firms in these markets has the potential to amplify market wide shocks and could cause significant disruption. In these markets, firms of all sizes, large to relatively small, provide crucial clearing and settlement services to other market participants and as such, could pose contagion risk if they fail. Several of the firms assessed are subsidiaries of systemic firms and, in certain stress scenarios, could pose severe liquidity and/or reputational risks to their ultimate parent and wider group.

2. Who will this interest

This review will be of general interest to all firms in scope of IFPR’s governance and liquidity risk management requirements, and to the UK parent entities of investment firm groups in scope of the IFPR.

The review allowed us to compare and contrast firms’ approaches to liquidity risk management across similar business models and also more broadly across a wide range of sell-side business models. The purpose of this publication is to raise awareness by sharing the findings of our review. We encourage firms to use this review including the good and poor practices section to improve their practical liquidity risk management capabilities and ensure that their framework is effective in identifying and managing the specific liquidity risks inherent in their business models.  

This publication is part of a broader communication process. We will organise workshops and roundtables with firms, industry trade bodies and consultants to share our observations and findings. We will also take questions and participate in practical discussions aimed at improving liquidity risk management in this sector and, encourage adoption of good practices.  

Nothing in this paper is intended as a change in FCA policy or rules. To understand our existing policy, you should refer to our rules and guidance on assessing adequate financial resources under the Prudential sourcebook for MiFID Investment Firms (MIFIDPRU). 

Please see final section for other papers that may be useful to read in conjunction with this review.

3. Summary of our approach

Our review covered twenty-six of the larger firms across the business models listed above that are prudentially supervised by the FCA. We focused on assessing firms’ practical approach to liquidity risk management. Our analysis incorporated information from several sources including the regulatory returns which we already hold, a detailed questionnaire, and meetings with firms to discuss the practical application of their liquidity risk management framework. In some instances, it also included a review of firm’s liquidity risk and Internal Capital Adequacy and Risk Assessment (ICARA) process documentation. 

We used this information to evaluate the effectiveness of each firm's framework, with a specific emphasis on business model, governance arrangements, risk identification, risk quantification, timing and escalation procedures, and operability of mitigating actions including contingency actions. We also assessed firms’ liquid asset threshold requirement (LATR) calculations, liquidity stress testing and operational arrangements.

4. Key findings and actions

We identified firms applying a range of approaches to managing their liquidity risks. In many cases, these approaches were appropriate and proportionate to the nature, scale and complexity of the firm’s business model. However, some firms had weaker approaches that were not commensurate with their size, complexity and the instantaneous nature of their liquidity risks. Often these firms had not updated their assumptions in the light of the events of the last few years. In general, we found that these firms:

  • failed to identify the full range of liquidity risks they are exposed to, especially idiosyncratic risk requirements
  • under-estimated the quantum of their liquidity risk exposures
  • relied too much on having immediate access to liquidity facilities to mitigate instantaneous liquidity requirements
  • had inoperable contingency funding plans (CFPs), many of which lacked action triggers, or a range of contingency actions designed to mitigate liquidity stresses

Compared to stronger peers, several firms had weaknesses in their approach to liquidity stress testing that left them vulnerable to severe market events, and unprepared to cope with moderate idiosyncratic events. We also found that the CFPs of these firms lacked a range of contingency actions to allow these firms to mitigate even commonly identified liquidity stress scenarios in a timely manner. These findings reinforce our message from our Dear CEO letters regarding a lack of experience and under-estimation of the severity of events.

All firms in our study identified intra-day (T0) and inter-day (T1) stressed cash outflows as their primary liquidity risk, with firms modelling - on average - 80% of their stressed liquidity outflows occurring on T0 or T1. We observed some common features in firms with well-functioning frameworks:

  • at close of business each day, firms dynamically calculated their immediate term business-as-usual (BAU) and liquidity stress liquidity requirements. Focusing their forecast on, T0 next day, T1 the day after that, and thereafter the short-term contractual and potential behavioural and stressed cash inflows and outflows
  • these firms promptly identified when a stress was likely to crystallise and assessed the likely impact
  • firms were prepared to take mitigating actions promptly in advance of a possible stress or, if that was not possible, in a timely manner as soon as they were impacted by a stress event - proactively forecasting and being prepared to respond to potential stresses; reinforced by monitoring trade flows and operational processes across the day and reacting promptly as anticipated and occasionally unanticipated issues emerge

Following our reviews, we took the following action:

  • All firms in scope of our review received direct feedback.
    • We identified weaknesses in 14 firms’ frameworks including underestimated liquid asset threshold requirements (LATR). For these firms, we set Individual Liquidity Guidance (ILG) and asked them to provide an attestation and assurance report to confirm remediation.
    • We identified weaknesses in a further 3 firms’ frameworks. We asked these firms to provide an attestation and assurance to confirm remediation.
    • We identified areas for improvement in 9 firms’ frameworks. These firms received feedback on areas of improvement.

Where we identified potentially critical weaknesses, we provided firms with prompt initial feedback.

We will continue to use these and other regulatory tools where we find firms are not properly managing their liquidity risks. 

5. Good and poor practice

5.1. Governance and risk culture

Good practice

  • A culture of managing risks primarily to optimise longer term performance and resilience, and secondarily to comply with regulatory requirements.
  • Dynamically, at the end of each day, identifying and quantifying their inherent forecasted intraday (T0) risk exposures, considering the effectiveness of their BAU and contingency mitigating actions and ensuring that their residual forecasted intraday and inter-day risk exposures remain within the risk tolerances defined by their risk appetite.
  • A clearly defined qualitative and quantitative risk appetite, which:
    • clearly set limits and action triggers around how much risk the firm is willing to take
    • is used to set the required levels of resource (financial and non-financial), and
    • is not solely based on adequate financial resources but also on non-financial criteria.  For example, some firms had action triggers based on systems outages, reconciliation failures, the nature of contingent liabilities and the financial position of the overall group
  • Ensuring their framework remains fit-for-purpose by frequently:
    • reviewing their framework with regular assurance reviews conducted by either risk oversight or internal audit, and
    • analysing internal and external risk events including near misses and incorporating the findings of their reviews into their framework
  • Continuously improving frameworks and operational processes with a primary focus on reducing intraday liquidity risk. For example, some firms had pre-emptively introduced a range of measures into their standard BAU processes to help them withstand instantaneous liquidity shocks.
  • A holistic approach to risk management with market, credit, operational and liquidity risk considered in a joined-up manner.

Poor Practice

  • Many firms had well designed frameworks but struggled to embed them.
  • In many instances, strong frameworks remained reliant on the ongoing sponsorship of one or more senior executives and were not a fundamental part of a strong risk management culture.
  • Some firms had not assessed whether their CCP membership category continues to be appropriate for their business model, scale and level of financial resources. An example of this risk crystallised in September 2018 when Nasdaq Clearing AB placed a member of its Nordic market in default. The losses were large and significantly exceeded the margin placed by the defaulter and the CCP’s own capital. This depleted the commodities default fund with the consequence that other members of the market needed to cover the excess losses within two business days. In the event, this sufficed to absorb the loss resulting from the default. However, not alone did the default force a CCP to pass on losses to members. Members were also required to replenish the default fund – the critical nuance of this default is that this extra step surprised some. 

5.2. Stress preparedness

Good practice

  • Conducting two forward looking liquidity stress assessments each day: one before banks and trading counterparties close; the other later in the evening.
  • Well considered stress assessments typically involved the following steps:  
    • Estimating the next day’s T0 peak BAU and potential stressed liquidity requirements (specifically margin calls).
    • Ensuring liquidity stress models included a broad range of operational inputs and behavioural analysis covering clients, counterparties, and liquidity providers. These models typically estimate peak potential outflows from margin calls, settlement failures or increased settlement requirements, client facility drawdowns, withdrawal of unsegregated client money, pre-funding requirements and failure to rollover debt.
    • Notifying group companies of large potential next day use of liquidity facilities including when above agreed limits.
    • Minimising the risk that liquidity facilities could be partially or wholly unavailable by ensuring at the end of each day adequate levels of liquidity to meet the next day’s instantaneous BAU and stressed cash outflows is placed:
      • in operational bank accounts
      • with CCPs such as LME Clear and ICE Clear; central clearing houses (CCH) such as London Clearing House (LCH) and Euroclear, and clearing brokers  

      To be as prepared as possible to cope with foreseeable large liquidity outflows.

  • Identifying large open positions, especially the short position legs of matched-principal trades, and ensuring inventory or stock borrows are in place so that these trades can be settled, avoiding the risk of being bought-in.
  • Ensuring that in certain defined circumstances clients have prepositioned cash/inventory for next day settlement.

Poor Practice

  • Stress testing at frequencies, such as monthly or less frequently, that did not reflect the instantaneous nature of liquidity stresses.
  • Stress testing that did not reflect the timing and magnitude of actual and potential liquidity outflows the firm has been exposed to in recent years. For example, during the nickel crisis firms:
    • underestimated the number and magnitude of margin calls (significant liquidity outflows) they had to meet promptly from CCPs for cleared derivative contracts across the day and into the next day, and
    • overestimated the timing and magnitude of margin (liquidity inflows) received, if at all, from their clients for ‘matched’ uncleared OTC derivative contracts, at some point during the next day
  • After the crisis, firms assumed the LME’s daily price-limits, implemented after the spike in Nickel prices, would prevent similar extreme price movements happening across several consecutive days (which they were not designed to do) rather than during a single day as happened in early March 2022 (which they were designed for).
  • Failing to calibrate value-at-risk (VAR) models, used to calculate their potential margin call liquidity requirements, to accurately forecast the potential timings and gross magnitude of each individual material cash outflow.
  • Failing to cross-reference lookback-period based calculations, used to quantify peak stressed liquidity requirements, with the firm’s current positions to estimate peak exposures based on current business levels.
  • Failing to include a realistic range of idiosyncratic stress scenarios and assumptions that would increase their liquidity requirements, reduce liquid asset levels or curtail access to sources of liquidity.
  • Failing to capture implicit credit exposures and liquidity usage arising from:
    • the time given to clients to meet OTC margin calls, and
    • naked short settlement positions and the consequences for the firm if it was bought-in, ie, the client(s) failed to deliver the inventory
  • Not considering maturity/roll-over risk impacting the effectiveness of medium-term notes, structured notes and similar programs. Despite the lessons of the global financial crisis, firms continue to rely on less stable funding sources to varying degrees and underestimate their exposure to:
    • roll-over concentration risks
    • auto-maturity concentration risks, and
    • risks of negative signalling by not being prepared to adequately support buy-back programs during periods of stress
  • Failing to consider and model the behaviours of clients specifically in an idiosyncratic liquidity stress. For example, clients withdrawing assets and cash held over-and-above what is required under title transfer collateral arrangements (TTCA) and reducing risk exposures which in turn further increases levels above TTCA requirements. Clients carrying out enhanced checks and reconciliations which increases delays on when payments are made, including margin payments.
  • Failing to adequately consider the behaviour of liquidity providers and not anticipating that:
    • parent companies or other group companies may not only severely curtail or withdraw liquidity facilities but also sweep liquidity back to group during stress events. Withdrawal of liquidity facilities and liquidity sweeps are common recovery and wind-down actions in group recovery and wind-down plans
    • third-party liquidity providers may severely curtail or withdraw liquidity facilities, either because they are in stress themselves or because they have reconsidered their risk appetite
    • for operational reasons, such as service outages at correspondent banks, access to liquidity facilities may be severely curtailed or totally unavailable

Recent financial stability board (FSB) reports have highlighted that the functioning and resilience of the non-bank sell-side ecosystem depends on the availability of liquidity and its effective intermediation under stressed market conditions. Certain activities and types of entities (so-called ‘key amplifiers’) are more likely to contribute to aggregate liquidity imbalances. Shocks can also be amplified due to the size and structural characteristics of such entities and their behaviour in stress. On the liquidity demand side, shocks manifest through unexpectedly large margin and collateral calls for derivatives and securities financing trades. This can give rise to asset fire sales by market participants seeking liquidity to cover these calls and the transmission of stress to other parts of the financial system and the real economy.

Finally, we observed that in their risk horizon scanning, firms were not considering emerging risks. For example, risks arising from technological innovations such as messaging applications and applications allowing clients to access their accounts remotely, thereby amplifying the impact of rapid behavioural responses such as withdrawal of money and assets, including surplus money and assets held under TTCA. This highlights the need for ongoing monitoring of all types of funding providers, including clients (especially those with credit balances), lenders, clearers, liquidity providers and providers of banking services so that firms may act promptly to make alternative contingency arrangements, where necessary.  

5.3. Contingency funding plans (CFP) and wind-down plans (WDP)

Good practice

  • CFPs with clearly defined quantitative action triggers that if breached provided Board delegated authority to a named individual to take action from a Board-approved list of contingency actions.
  • Clearly identified contingency actions, appropriate to each firm’s unique circumstances, that could be taken promptly:
    • For example, actions to increase liquidity levels such as:
      • drawing-down from liquidity facilities
      • temporarily increasing the size of liquidity facilities
      • selling unencumbered inventory
      • leasing out precious metals
      • using sale and repurchase (repo) arrangements to preclude outright sales crystalising losses, and/or rehypothecate collateral held 
        Being mindful to avoid negative signalling.
    • For example, actions to conserve liquidity such as:
      • reducing trading limits
      • reducing size limits for transactions in single-name securities/commodities
      • having clients pre-fund large and/or illiquid transactions and having clients pre-position inventory (securities and/or commodities) to cover sales or provide proof that executable stock borrow arrangements are in place
      • using name give-up and exchange give-up of unsettled transactions
      • having clients pre-fund potential variation margin requirements arising from larger derivative positions or refusing credit facilities to clients to cover larger back-to-back margin calls
      • reducing the amount of credit provided to clients. For example, to ensure this action can be taken promptly firms:
        • periodically updated the list of impacted clients and the amounts by which credit facilities would be reduced
        • considered and modelled anticipated client behaviours in range of stress scenarios
        • considered the likely consequences of taking this action
        • included in their contingency plans agreed client communications strategies and, where appropriate, coordinated communications strategies across the wider group.
  • Frequently test the operability of contingency actions during business-as-usual conditions and avoid the risks of negative signalling that arise if actions are only implemented during a stress situation. For example, frequent random drawdowns from committed facilities and increased utilisation of uncommitted facilities.   
  • Considered in their WDPs the outputs of reverse stress testing (RST), ie stresses that would lead to a point of non-viability and result in the WDP being invoked.
  • Ensuring WDPs are operable in a stressed environment and included contingencies for activities that potentially require longer execution times. For example:
    • notice periods required to terminate contractual arrangements
    • time to provide notice to clients so assets can be migrated off the platform
    • the eventuality that some clients may not be immediately contactable
    • resolution of issues that would arise regarding disputed client money and client asset positions, delayed and/or disputed settlements, and aged open reconciliation issues
  • WDPs which include projected profit and loss, balance sheet and cashflow statements at appropriate levels of granularity. For example, at the beginning of the process when multiple critical cash outflows and inflows are expected to happen within similar timeframes, the cashflow analysis was presented by day/week detailing the timing of all material cash inflows and outflows.  

Poor Practice

  • Failing to consider or put in place CFPs that can be invoked and executed in a reasonably timely manner in a period of stress.
  • Not practicing, not being able to implement theoretical steps.
  • CFPs without clearly defined quantitative action triggers, especially non-financial action triggers. For example, during the cyber-attack on ION’s cleared derivatives platform in 2023, many firms continued to trade at normal volumes even though their middle and back office were no longer operable. Consequently, trade volumes materially exceeded their capacity to adequately reconcile their positions.  
  • Failing to consider the full range of actions available. Even though other actions were available, several firms included only one contingency action, usually to draw-down from a single liquidity facility, or request funding from their parent. Firms that have not considered the full range of contingency actions available to them would be at greater risk of failure in the event of a sudden liquidity stress. 

5.4. Liquidity risk management capabilities

Good practice

  • Teams with appropriate levels of expertise:
    • constantly updated their expertise and knowledge
    • constantly upgraded their frameworks to meet the ongoing realities of their businesses
    • put processes in place that act in tandem with the firm’s operational processes to help the firm effectively manage market and idiosyncratic liquidity risks and cope with periods of extreme market volatility and idiosyncratic shocks such as severe operational risk events
  • Ensuring a holistic approach to liquidity risk management. For example, having a clear understanding of the timings and magnitude of BAU and stressed cash inflows and outflows arising from operational processes, especially outsourced operational processes.

Poor Practice

  • Teams responsible for liquidity risk management lacking expertise in:
    • the liquidity risks inherent in the firm’s business model, leading to these risks being understated or absent from the liquidity risk stress testing model, CFP action triggers and contingency actions
    • the operational processes of the firm and the activities of the operations teams.  Where this happens, frameworks are not designed to support the activities and processes of the operations team. For example, some firm’s liquidity risk managers could not promptly identify when operations, processes or controls had failed and therefore could not promptly alert the treasury function of the need to take mitigating actions
  • Weak understanding of outsourcing arrangements especially where separate unregulated subsidiaries provide operations and related support services. From a practical perspective failing to ensure that the liquidity and funding risks arising from these outsourced activities are identified and proactively managed with appropriate oversight either at a group level or at the solo regulated subsidiary level. And, not including assessments of these risks in either the group ICARA or the ICARAs of the regulated subsidiaries.
  • Applying a tick-the-box approach to liquidity risk management. Some firms’ frameworks were narrowly designed around meeting regulatory requirements rather than being designed to be ‘fit-for-purpose’ in line with the practical principles of sound liquidity risk management.

6. Data quality – inaccurate or incomplete data submissions

Our prudential planning and approach rely more and more  on regulatory data. Responses based on inaccurate and/or poor-quality regulatory data submissions creates a negative loop of extra work as we probe firms’ responses and may undermine the effectiveness of responses in the event of market events.

Firms providing inaccurate or incomplete data may be in breach of their responsibilities under the senior managers and certification regime (SM&CR) and Principle 11 which require firms to comply with the relevant requirements and standards of the regulatory system and disclose any information of which we would reasonably expect notice.

Our expectations related to data accuracy have been outlined in previous ‘Dear CEO’ letters on Transforming data collection (February 2021) and Quality of Prudential Regulatory Returns (February 2018).

7. Useful papers to read in conjunction with this review