The implementation of credit controls are important; however, they should not lead to discriminatory execution or remove access for a segment of the FX markets. Given appropriate tools for measurement, monitoring, and enforcement of credit, FX intermediaries can effectively manage their exposures to clients and liquidity providers without wholesale disruption to existing global execution models or the credit processing and framework that underpins the FX markets.
The industry does not need to tear-up existing technology and tools, procedures, and controls in order to achieve the desired balance for all FX participants; however, engagement and involvement in defining core API, data, and interactions is a crucial first step. Over the last 20 years, the evolution in global FX execution platforms, products and the legal and regulatory framework has led to enormous growth in FX Markets but has created numerous challenges. Unlike other asset classes with centralized venues and exchanges, FX is a hugely fragmented marketplace with multiple venues and liquidity pools. The unbundling of execution and credit has fueled the rise of non-bank market makers and algorithmic or systematic trading across buyside FX participants. Credit intermediaries have largely succeeded in managing the risks for G10 Spot, however we are now seeing greater electronification in other FX products- NDFs; FX Swaps; FX Options and EM currencies resulting in increased credit challenges.
Although market-driven disruptions have been few relative to the scale and volume of FX execution, their cause and potential long-term impact recently has led to calls for widespread changes to the existing risk models, methods, and controls.
In this paper, we propose principles for a credit utility model which addresses some of the challenges faced by FX market participants. We will do this not only by looking back at some market disruption events, but also looking forward to forthcoming challenges within the FX industry to assess both the strengths and weaknesses of the current process and framework.
Our utility credit model will identify some potential areas of focus to improve the current framework without destroying access to liquidity. Importantly, the desired evolution cannot be achieved in isolation, a partnership across FX markets participants, vendors and industry bodies would be beneficial for a viable end state.
Credit is the fuel for execution liquidity and must be controlled pre-execution until full and final settlement. In the FX OTC execution model, all aspects of an agreed trade — legal, credit, market and operational (confirmation/settlement) — are typically dealt with and managed directly between the two transacting (executing) parties or via a credit intermediary, normally a prime broker or prime of prime. While vendors and/or third parties may assist in supporting these post-trade processes, as shown in Figure 1, the original trading relationship and intermediary is crucial.
How can intermediaries control client exposures without disrupting the price discovery process? FX is a hugely fragmented marketplace with multiple venues and liquidity pools, this is unlike IRS, CDS, ETD, or equity processing, where central clearing is mandated or execution occurs on an exchange under an open offer model3. Exposures and margin are typically reflected against a central counterparty (CCP) who manages the risk and ‘insures’ all parties against the default of a single entity. However, where execution occurs off-exchange or via third-party brokers, the OTC/ETD intermediaries are facing similar challenges to FX OTC.
In FX, the bank and prime broker intermediaries play a crucial role for all participants in ensuring that FX execution can occur on multiple platforms, but more importantly, settlement finality is achieved by all executing parties. FX credit is defined, agreed to, and controlled through legal agreements. Figure 2 shows the timeline view of FX credit risk; in its raw form, it can be broken down into execution risk, counterparty risk, and settlement risk.
Firms use well-established legal frameworks ─ bilateral , tri-party4 or post-trade clearing agreement (since 2012) ─ to underpin the FX ecosystem and help manage and control credit risks for all participants.
Execution, counterparty, and settlement risks arise from the point an FX order is submitted by the price taker. Credit intermediation allows the price maker to provide a price(s) and enable execution. Once the relevant matching process or RFQ5 completes and the order is filled, the taker and the maker face counterparty risk to each other. Where a third-party steps in for either party, the post-trade give-up process transfers the relevant exposures based on legal agreements. If a buy-side client is involved, the parties may need to allocate trades into the respective end accounts to reflect exposures into relevant funds. On the value date of the contract, firms may face settlement risk, as both of the pay and receive amounts may not be exchanged simultaneously and one party could fail in its final obligation.
Importantly, all participants in the ecosystem are involved in using, controlling, or managing credit risks to enable a smooth functioning ecosystem:
Any proposed changes to the existing FX credit risk framework will need to consider implications for the pre-trade to settlement finality risks but also the potential negative impact on market access - Without all participants’ involvement in the end-to-end process, FX execution microstructure would not have evolved to become the largest and most efficient market. Indeed, the current model of participation is inclusive, diverse and efficient.
Central banks have a vested interest in ensuring a safe and efficient FX marketplace, both in support of their macro policy perspective but also for local currency stability - One unique dynamic of the FX markets is the important role it plays in enabling a well-functioning local economy and in supporting the local central bank’s policy objectives. These government agencies regulate, monitor, influence but also actively participate in the FX market micro-structure. However, due to the global nature of FX, the local policy needs may sometimes conflict with a well-functioning global marketplace.
FX credit processing has evolved through innovation and investment in technology. Commercial opportunities and need for both in-venue controls and post-trade mitigation tools have allowed participants to actively engage in diverse liquidity pools. The unbundling of execution from credit processing has enabled diversity of both execution participation (non-bank, buy side systemic trading) and credit intermediation (FX Prime Brokers and Prime of Primes).
Products such as spot are much more automated in end-to-end execution to settlement processing while others (e.g., FX options) may be high touch. Figure 3 represents the typical process and in some cases the interactions may be more manual, depending on the execution method or platform. The trends to date have largely been in spot- but over time we would expect greater electronification in other FX products- NDFs; FX Swaps; FXO and EM currencies resulting in increased credit challenges.
Some platforms operate execution matching exclusively for bank-to-bank liquidity as they are the recipients of bilateral credit. Other platforms are focused on client-to-bank activity due to their access to a central PB or multiple PB credit lines - With the diversity and growth of execution venues, many of the process touch points and services are relationship and network dependent (connectivity and technology). Indeed ‘credit’ provided by the bank or PB intermediaries is a differentiator for the execution venue, without this relationship the venues would not be able to offer execution services.
Pre-Trade: In-venue & ping (execution risk)
Post-Trade (Counterparty & Settlement Risk)
Along with defining the end-to-end process, bank and PB intermediaries have developed a standardized framework of tools to enable robust risk management. Figure 4 highlights four key pillars associated with second line risk principles: manage, measure, monitor and control.
As with most frameworks, there are many challenges with aspects of the toolkit. Before assessing potential improvements, we will examine why FX credit risk has become a focus for all participants and the central banks in particular.
Our comprehensive suite of services allows you to manage both credit risk and legal documents to set, amend, and monitor your clients’ exposure in real-time.
Download Credit Risk factsheet
Turkish lira events will provide focus areas for what needs to change with the FX credit process or framework. Many studies and articles been written about the high-profile events that have led to structural changes in the FX markets microstructure. The extreme events of failure of Herstatt bank in 1974 led to the creation of CLS8 in 2002. The 2015 Swiss franc currency de-peg event led to market volatility and bankruptcy (Alpari UK). There is also the more recent structural shift to Emerging Market (EM) currencies and growth of importance non-bank financial and corporates. Each of these events provided opportunities to reflect and improve upon existing FX market microstructure. We will not cover each of these events in detail but instead will focus on the recent events relating to Turkish lira, as this encompasses most of the current major structural themes. Our ultimate assessment will examine where and how either the FX credit process or the framework failed to provide enough safeguards or controls for both the credit intermediaries and participants.
The long-term implications are crucial to understand for a viable process and framework. In addition to looking in the past, we will focus on some potential forthcoming regulatory challenges to assess how the current process and framework could be future-proofed given the movement within the financial regulatory landscape.
Turkey’s desire to be included in the European Union led to open capital markets and development of financial market infrastructure to help support growth in trading of local equites and bonds (both government and corporate), which attracted not only asset managers, but also speculators seeking greater opportunities. Over the last few years, the Turkish lira has sustained a period of uncertainty due to local economic conditions and government policy. Regardless of the root cause, we have seen multiple currency crises starting from as far back as 2002, and the issues have been exacerbated over the last few years due to the dynamic changes in participation for Turkish Lira assets.
Currency markets can provide an efficient mechanism for local and foreign participants to deal with local funding needs, however, intraday funding gaps, government intervention, and policy uncertainty can create risks and issues. Importantly, unlike many emerging markets, the lira is a deliverable currency, but due to settlement occurring outside of CLS infrastructure, there is a reliance on local correspondent banks for physical settlement. This nuance means that foreign participants can and do utilize local entities to manage currency balances and visa-versa, typically on a foreign exchange venue. This implies that all participants are therefore subject to settlement risk associated with these local participants and may in most cases require credit from intermediaries, who in-turn will rely on the local banks for settlement.
The combination of economic uncertainty, policy changes, and requirement for real-world physical settlement of currency created a squeeze on funding markets, which led to large volatile moves in FX rates and subsequent triggers for OTC FX options, a perfect storm for credit intermediaries - Additionally, due to greater access to technology and the hunt for alpha, non-bank firms gravitated towards Turkish lira execution both for assets (bonds, equities) and for currency trading. Non-bank participants can provide liquidity both as makers and takers in spot. They normally rely on intermediaries not only for credit to execute but also for settlement purposes. This in itself is not a bad situation and can be observed, efficiently across many other emerging markets currencies as long as firms have the ability to execute and settle FX in an efficient manner.
It could be argued that these events and circumstances are unique to Turkey and are an anomaly; however, we have seen similarities with currency shocks in Russia, Indonesia, and will continue to experience these going forward with Argentina, Poland, China, India, and Brazil which are volatile currencies.
1. Volatility and instability with settlement of lira – While no large losses can be directly attributed to the volatility in lira, participants have faced challenging execution and settlement environment when the Turkish government has tried to control or stifle offshore markets by limiting local banks’ participation.
Traditional net open position (NOP) and daily settlement limits (DLS) measurements may not recognize risky or volatility settlement conditions in an appropriate manner. Better credit controls across a wider spectrum of venues and intraday liquidity savings mechanisms or pre-settlement tools and funding may be beneficial or required where CLS processing is not available. The presence of foreign participants in a local market can be beneficial to growth of the local economy, but in some cases it can cause short- to medium-term instability if the country does not have a strong banking environment to support both the supply of currency for execution and settlement. Most traditional Emerging Markets operate a closed and segregated on-shore market and a separate off-shore market. However, the Turkish government has partially implemented their liberalization program, allowing local participation in offshore market and government-controlled ecosystems with limited flow of capital both in and out. For credit intermediaries who facilitate both credit and settlement, this creates problematic execution controls and settlement environments due to the dependency of available supply of currency from local banks.
2. Large PB suffers losses from client FX options exposures – In 2018, PB reports of losses in the range of $180m vs $30m in margin against a single counterparty due to exotic FXO.
Firms offering FX options should ensure relevant margin models that are linked to credit measurements. Additionally, management of inventories is crucial; otherwise, pin risk can cause mismatch in settlement exposures that cannot be met due to stickiness in local funding markets. FX options are widely used by many client segments in the global economy to effectively manage funding needs or to speculate. Corporates will utilize FX options to lock in or trigger payoffs only if certain FX rates are prevalent but ultimately, they need the foreign currency payment (PvP). Speculators will use FX options as a cheaper way to implement a specific strategy. Importantly, margining and collateral controls are essential and should be linked into credit for execution purposes, but due to the nature of options exercise and post-trade settlement, the risks in FX options can be more extreme than cash products. In this specific scenario, the PB would have had a matched book, i.e., client->PB and PB->EB, and potentially because of this model, the intermediary may not have had sufficient tools to manage the risks appropriately. It was evident that the PB margin requirement was well below the potential losses.
G10 policy goal is to incentivize central counterparty clearing and reduce the overall systemic risk in OTC markets by imposing margin and balance sheet (leverage and liquidity) constraints, while the FX Global Code is a set of standards that FX participants have voluntarily adhered to. Since the financial market crisis of 2008 and subsequent Pittsburgh 2010 agreement, global regulatory authorities have focused on strengthening the financial markets and curbing excessive risk within the OTC markets. The impact to FX markets has largely been muted to date due to the cash nature of the product, but also, it’s importance in financial markets as a tool for global commerce and macro-economic stability. Over the coming years, four areas of the regulations may impact FX business models and its participants.
Margin: The bilateral margin rules imposes two-way initial margin (IM) and variation margin (VM) requirements on non-cleared OTC. For FX, the implementation is product specific and the impact is dependent on region and size of institution:
Unlike IRS and CDS, there is no mandatory clearing obligation for FX products; therefore, executing parties and intermediaries have a choice. This implies that FX credit models may need to incorporate multifaceted execution to settlement paradigms, including PB to clearing, bilateral to clearing, or a cleared product from point of execution. Due to the nature of the FX cash business, a large portion of the FX market will not be directly impacted by the margin rules. However, derivatives products, mainly NDFs and FX options, will fall in scope, and while the full implementation timeline will only come into effect by Sept. 2022, the implications could require changes across most firms’ operating models. Costs pressures for credit intermediaries, both for the dealer banks and prime brokers, imply that most if not all NDFs would likely be cleared in some form. As of March 2021, the majority of the dealer-to-dealer NDF flow is being cleared in order to reduce margin impact; however, very little client activity is being cleared. PBs in some instances are clearing the EB->PB leg as this leg is captured under the bilateral rules for both IM and VM; however, the client->PB leg remains OTC since most clients are not impacted until Sept. 2021 or Sept. 2022. Over the next few years, most firms will assess operating models and look to reduce the direct burden of margin for both FX options and NDFs.
Execution venues and optimization providers will look to enhance margin management tools, credit process, and framework to support both deliverable and non-deliverable G10 products. Product innovation is also very likely. Due to margin requirements on FX options but not on FX spot and forward hedges, some firms may look to execute non-deliverable G10 hedge trades in order to manage their bilateral margin requirements.
Leverage: The implications are that margining will become more relevant and important in managing balance sheet exposures, even for deliverable products which do not require posting of IM or VM. In addition, larger firms may be incentivized to reduce their overall exposures facilitating growth in different forms of credit intermediation (prime of prime) and non-bank liquidity providers. The supplemental leverage ratio (SLR) is intended to reduce the excessive bank balance sheet usage that was observed prior to 2008. Globally, systemic important banks (G-SIB) have an add-on to SLR, this is intended to curb firms from becoming too large or significant in any given sector or asset class. Although detailed analysis is out of scope for this paper, the key driving factor for FX participants will be the transition from the current exposure method (CEM) to the standardized approach to counterparty credit risk (SA-CCR) methodology. In general, the SA-CCR method is a more risk-based approach as it recognizes margin vs. exposures and offers greater netting opportunities. Firms may find it more efficient to trade FX swaps (smaller net exposure) vs outright spot or FX forwards, as the netting impact could be significant in its balance sheet usage.
Liquidity: Due to focus on correspondent banking risk associated with intraday liquidity exposures, over the next 5-10 years, the industry may see growth in central bank digital currency (CBDC), implying greater direct access to central banks settlement, liquidity savings mechanisms, and potential for product innovation including execution and PvP of shorter-dated FX digital products. Given that FX is systemically important for funding markets, a lot of attention has been put on short-term liquidity, both with the liquidity coverage ratio but also through the implementation of intraday liquidity management, measurement, and reporting. Although no direct reserve is held against intraday exposures at this time, it is anticipated that regulatory authorities will focus more on firm’s ability to be able to manage intraday funding gaps arising from settlement of FX and other trading activity.
Of interest will be the correspondent banking risk that most firms take when settling foreign exchange transactions outside of CLS. Due to limitations and opening time differentials with local real-time gross settlement systems (RTGS), settlement firms do extend large amounts of currency intraday to enable a smooth-functioning, wholesale, global bilateral settlement system. We are, however, observing slow evolution in this area of the financial markets. Swift is in the process of implementing better visibility and management of end-beneficiary risks, however, the core processing still relies on extension of credit. The possibility for instantaneous digital settlement across a closed ledger is being explored by technology disruptors and various bank consortiums.
FX Global Code of Conduct (FXGCC) - Initially published in 2017 with many revisions since, the FXGCC has been developed to provide a set of common guidelines to help promote effective functioning and integrity within the wholesale foreign exchange market (the FX market).
Over 1,000 participants have signed a “letter of intent,” and the FXGCC is explicit on what principles need to be incorporated for firms who participate in wholesale FX markets; however, it does not define how and what needs to be implemented. If the FX markets do not take the lead, future iterations may include explicit regulations. The core principles cover ethics, governance, execution, information sharing, risk management, compliance and confirmation and settlement process. The principles of most interest for FX credit processing and framework involve last look, best execution, greater automation for FX STP, and, importantly, focus on credit provisioning controls for FX intermediation activities.
While our assessment is in no way all-encompassing, we have been able to provide visibility across some areas of the process and framework that require improvement, but also viewed some potential challenges that will need to be incorporated into any strawman. Crucially, the FX markets and its participants can control how and what changes are made to improve the current situation.
SEF and MTF regulation could be perceived as onerous for participants and venues; however, they have allowed for a central credit utility model to evolve around a functional process and framework. Although not covered in this paper, the cleared OTC markets have experienced the consequences of what a regulated rules-based credit process and framework could imply to execution operating models. The CFTC SEF regulations (1.73/1.74)9 and ESMA’s MTF Articles (RTS 26)10, have imposed a very rigid order screening and credit framework for cleared OTC derivatives (IRS/CDS/FX) across regulated venues.
Technology is a key enabler to resolving some of the improvements we have highlighted, but in many ways, defining industry standard business process and data standardization will be essential to achieve a viable end state.
In the below we consider a strawman utility model that fits the needs of the FX market participants. We focus on core principles that would allow us to address the issues and needs of the market participants, including central banks.
Credit utility should allow and enable independent evolution and growth of execution liquidity pools while providing all participants with visibility, control, and most efficient use of credit - An industry utility model to link in execution to full and final settlement processing across a more diverse set of FX liquidity pools. Enabling and extending the current credit process and framework while adhering to the core principles.
Designing a process model in partnership with industry body and representation from all categories of participants is a crucial first step. Initial engagement would need to involve a partnership with an FX industry body. All primary participants would contribute to the definition of a set of core data models and APIs that would allow participants to interact more efficiently but also establish some core de minimis requirements for each participant in their interactions with the FX Credit microstructure.
Although not mandatory, a central utility service could be the centerpiece of the ecosystem and could implement and enable the interactions. The merits and benefits of a utility model are well understood with-in the financial markets and provide value for both participants and regulators. We have observed this with varied level of success within the FX markets with the introduction of CLS in 2002. FX clearing infrastructure has evolved organically since early 2010s, with some market penetration for the NDF product but little traction across other cash products. Importantly, the challenges with an FX credit utility are much broader in scale and participation, unlike CLS and CCP, an initial bank-only focus may not address the core issues. The importance and inclusion of all participating firms ─ vendors, platforms, price takers, and makers ─are crucial to the successful transition into a viable FX credit microstructure. As a minimum, a holistic design approach should be taken in order to meet the needs of all participants in the ecosystem.
It is unknown if the utility would need to register as a credit financial market infrastructure (FMI) – for debate – however, an appropriate governance structure, rulebook, and legal framework would ensure appropriate access for FX participants. Additionally, future iterations could enable extension into other asset classes such as exchange-traded derivatives, repo, and OTC equity derivatives where similar credit intermediation issues exist.
While providing access and implementing core processing would be paramount to the service, the credit utility would need to ensure some core improvements to the current toolkit which would allow intermediates to meet their goals in order to manage, monitor, measure and control FX credit risks. Increasing participation, improving tools and links, increasing distribution, and addressing the overallocation issues would be the primary goals of the service.
Manage pre- and post-trade credit checks and monitor credit utilization across FX, IRS, and CDS.
Book a demo
The above provides a high-level view of what functionality would need to be encompassed into the utility service. Although much of the detailed picture will already be known to participants and providers alike, many new facets could be introduced with engagement across a wider spectrum of FX market participants. Technology is one of the many key enablers to finding a path forward, however the journey of migrating the existing ecosystem and network to a viable new FX credit framework and process must be a well-planned, gradual and inclusive evolution. Without this approach we risk destroying access for some participants and the growth that has been achieved over the last 20 years in the FX Markets.
Join Kevin McPartland, Greenwich Associates and a panel of experts from BNP Paribas, J.P. Morgan, Citi, and Traiana in a discussion on how the over-allocation of credit risk in FX impacts the largest prime brokers and other market participants.
In this whitepaper, we have tried to map out the story board for FX credit, by looking at what exists today against some of the past and future challenges. We hope the reader has been given an overall picture of the dynamics at play within the FX credit microstructure.
We explored some of our thoughts regarding a strawman for a credit utility model, highlighting the need for an initial industry driven structured approach to standardization, both in data and the need for a well-defined interaction model for participants. Importantly, we are not an advocate for or against a single provider of the utility service; however, some formal governance structure could be a useful characteristic for a utility service. Along with providing the core processing, we have identified some areas of improvement that the utility could provide to enhance the existing framework and toolkit around managing, measuring, monitoring and controlling FX credit risks.
Structured formal engagement is a first step and one of the core initial debates would center around the need for and merits of a central utility. Indeed, the intention of the paper was not to provide all the answers and disclose a definitive service proposal. Instead, this is the start of a journey. We welcome and encourage broader engagement and debate on the merits of extending the current process and framework to a central utility model in order to enable an efficient well-functioning FX credit ecosystem.
1 ISDA MA: An ISDA Master Agreement is the standard document regularly used to govern over-the-counter derivatives transactions. Credit Support Annex regulates collateral under the. ISDA MA by defining the terms and conditions under which collateral is posted to mitigate counterparty risk
2 PvP: A settlement mechanism that ensures that the final transfer of a payment in one currency occurs if and only if the final transfer of a payment in another currency or currencies takes place.
3 Open Offer: A process through which a CCP extends an "open offer" to act as counterparty to market participants and thereby is interposed between participants at the time a trade is executed.
4 Tri-Party: Terms for documenting foreign exchange "give up" relationships, in which a party designated by a prime broker executes transactions with a dealer that are "given up" to the prime broker. Multiple legal arraignments represent these relationships including, PB Master Agreements, Master Give-up Agreements,3-way, 4-way and Prime of Prime
5 RFQ: Request-for-Quote is an execution method where a quote(s) is provided by a participant in response to a request for a quote submitted by one or more other participants, the quote is a firm executable price exclusively available to the requesting participant
6 Streaming RFQ: The execution method permits an RFQ requester to request streaming prices for a given product from multiple participants, where those recipients are able to update their streaming prices according to regular intervals. The quote is normally indicative and may be subject to the platforms Last-Look policies
7 CLOB: Central Limit Order Book is an all-to-all market in which all liquidity in the order book is firm. In FX this will generally be available for inter-bank activity (with CLS settlement) or where a Central PB model is available as the post-trade facilitator
8 CLS: Continuous Linked Settlement service started operating in 2002 as an Edge Act Corporation, a limited purpose bank regulated by the Federal Reserve Bank of New York.
9 CFTC SEF regulations: below link is a guidance and high-level overview https://www.cftc.gov/sites/default/files/idc/groups/public/@newsroom/documents/file/stpguidance.pdf
10 ESMA RTS 26: Link to guidance and high-level overview https://ec.europa.eu/finance/securities/docs/isd/mifid/rts/160629-rts-26_en.pdf
Basu Choudhury works at Traiana, which is part of CME Group. He focuses on strategic service capabilities for both the buy- and sell- side firms. Basu brings a wealth of experience as the former head of product for LCH ForexClear service, where he lead the design of the service for FX NDF clearing and was at the forefront of the definition of a model for CCPs to interact with CLS to enable the clearing of other FX products (FX Options/Spot/Forwards). Prior to this, he spent over 15 years in roles involving business and process reengineering - building, designing and supporting banking and hedge fund business initiatives across Rates, FX, Equity Derivative and Credit products.
Basu received his Masters in Economics, specializing in econometrics and options theory, from the University of Toronto in 1997.