Skip to content

Cases

The authorities, and the questions they leave.

Vanilla and barrier options first — one-touch authorities at the head of that list — then the rest of the FX book. Public judgments and decisions. We were not instructed in them. Illustrative studies of the live questions follow.

Authorities

Reported cases

Public proceedings that set the market questions this practice is built for. Not a list of appointments. Current and former employers appear as parties in some of the files; that is disclosed at conflicts.

Vanilla and barrier options

One-touch first — the American digital that pays if spot ever trades the barrier. Then vanillas, then knock-in and knock-out barriers.

One-touch barriers

3 authorities

  • 2023–25

    S.D.N.Y. / 2d Cir.

    United States v. Neil Phillips (Glen Point Capital / Morgan Stanley)

    155 F.4th 102 (2d Cir. 2025); 690 F. Supp. 3d 268 (S.D.N.Y. 2023)

    A one-touch digital barrier option on USD/ZAR, barrier 12.50, sold by Morgan Stanley. Glen Point’s co-CIO convicted of directing some $700 million of spot to touch the barrier and collect a $20 million payout. The Second Circuit affirmed: what a one-touch is, whether the print was a good barrier, and whether chasing it was a market or a fraud.

  • 2022–

    S.D.N.Y. (CFTC)

    CFTC v. Glen Point Capital Advisors LP

    No. 1:22-cv-10589 (S.D.N.Y., filed 15 December 2022)

    The civil companion. Two one-touch binary options on USD/ZAR, combined payout $30 million, alleged to have been chased on 25 and 28 December 2017 in thin holiday liquidity. What the confirmation required as a barrier event, whether the print was a market, and the 1LOD/2LOD question of how a dealer records a one-touch that has just paid.

  • 2013

    Commercial Court

    Deutsche Bank AG v Sebastian Holdings Inc — digital and dual-range digital options

    [2013] EWHC 3463 (Comm)

    The same book. Digital currency options, dual-range digitals, knock-out timing options and fade-in forwards. Cooke J on whether a one-touch or dual-range digital is a “currency option” under the FX prime-brokerage agreement, or an exotic outside authority — and how a dealing room values a digital that has, or has not, touched.

Vanilla and exotic options

2 authorities

  • 2013

    Commercial Court

    Deutsche Bank AG v Sebastian Holdings Inc

    [2013] EWHC 3463 (Comm)

    Close-out of an FX and currency-options book. Cooke J on the market meaning of a vanilla put or call, and whether knock-outs, double knock-outs, digitals and target-profit forwards were currency options or exotic transactions outside the trader’s authority.

  • 2013–16

    Commercial Court / Court of Appeal

    Goldman Sachs International v Videocon Global Ltd

    [2013] EWHC 2843 (Comm); [2016] EWCA Civ 130

    Sums due on the termination of two vanilla currency option transactions (INR/USD and BRL/USD) under a 1992 ISDA. Whether the close-out followed the Master Agreement — quotes for spot, forwards and FX volatilities, and the options model used to turn them into a number.

FX barriers

2 authorities

  • 2013

    Supreme Court of Korea

    KIKO knock-in knock-out currency options

    2011Da53683 (en banc), 26 September 2013

    Knock-in knock-out USD/KRW options sold to Korean exporters. Whether the structure was an unfair contract, whether it was a hedge, and the bank’s duty to explain the knock-in, the knock-out and the unlimited downside.

  • 2010

    Commercial Court

    Titan Steel Wheels Ltd v Royal Bank of Scotland plc

    [2010] EWHC 211 (Comm)

    A ratio / accrual FX product with a protected rate and a barrier rate. Whether the bank advised, whether a duty of care arose, and whether the close-out of a structured FX trade with a barrier was a market.

Market conduct, close-out and competition

The rest of the FX book: last look, the tape, ISDA close-out, and the cartel files.

Close-out and ISDA

5 authorities

  • 2011

    Chancery Division

    Anthracite Rated Investments (Jersey) Ltd v Lehman Brothers Finance SA

    [2011] EWHC 1822 (Ch)

    Automatic early termination of cash-settled put options under ISDA after Lehman’s default. How the Early Termination Cash Settlement Amount is to be determined — a dealing-room close-out, not a screen mid.

  • 2012

    Court of Appeal

    Lomas v JFB Firth Rixson Inc

    [2012] EWCA Civ 419

    Section 2(a)(iii) of the ISDA Master Agreement after an Event of Default. Whether payment obligations are suspended, and what a close-out under the 1992 and 2002 forms is meant to achieve.

  • 2013

    Court of Appeal

    Lehman Brothers International (Europe) v Lehman Brothers Finance SA

    [2013] EWCA Civ 188

    Close-out of back-to-back inter-company transactions on a hybrid 1992/2002 ISDA. Market Quotation, Loss, and the “value clean” principle on termination of a derivatives book.

  • 2018

    Commercial Court

    Lehman Brothers Special Financing Inc v National Power Corporation

    [2018] EWHC 487 (Comm)

    Close-out of a USD/PHP forward currency swap under the 2002 ISDA. Whether the non-defaulting party’s Close-out Amount had to be objectively commercially reasonable, and whether a later, larger calculation could replace the first.

  • 2025

    Chancery Division

    Conway v Plass (Re Argentex LLP)

    [2025] EWHC 3125 (Ch)

    Administrators of an FX and payments firm sought to close out a book of forwards and options before maturity because the book was net positive. The court refused: the contracts were for certainty of a future price, not an early unwind for the firm’s own account.

Market conduct

6 authorities

  • 2014

    FCA

    Final Notices — Citibank, HSBC, JPMorgan, RBS and UBS

    12 November 2014

    G10 spot FX: ineffective controls allowed traders to share confidential client information and to attempt to manipulate WM/Reuters and ECB fixes, including in collusion. Fines totalling £1.1 billion.

  • 2015

    FCA

    Final Notice — Barclays Bank Plc

    19 May 2015

    The same G10 spot failings at Barclays’ London FX business: information sharing, attempted manipulation of fixes, and attempts to trigger client stop-loss orders. Fine of £284 million.

  • 2015

    NYDFS

    Barclays — Last Look on BARX

    18 November 2015

    Automated last look used to reject client orders that would have been unprofitable after a milliseconds-long hold. When clients asked why fills were rejected, the bank cited technical issues. Additional $150 million penalty.

  • 2017–19

    E.D.N.Y. / 2d Cir.

    United States v. Mark Johnson

    945 F.3d 606 (2d Cir. 2019)

    HSBC’s global head of FX cash trading convicted of wire fraud over a $3.5 billion Cairn Energy sterling purchase. Pre-hedging and ramping into a fixing, and whether the client was told the truth about how the order would be worked.

  • 2019–22

    S.D.N.Y. / 2d Cir.

    United States v. Akshay Aiyer

    33 F.4th 97 (2d Cir. 2022)

    A JPMorgan CEEMEA FX trader convicted of conspiring to fix prices and rig bids in emerging-market currencies. Coordination of bids and offers, withholding of supply, and concealment over chatrooms and personal phones.

  • 2018–23

    S.D.N.Y.

    Allianz Global Investors GmbH v Bank of America Corporation

    No. 18-cv-10364

    Opt-out funds alleging manipulation of the FX market, including last look on single-dealer platforms. Claims narrowed on the motion to dismiss; remaining claims resolved in 2023.

Competition

5 authorities

  • 2015

    D. Connecticut / DOJ

    United States v. Citicorp, JPMorgan, Barclays and RBS

    Parent-level guilty pleas, 20 May 2015

    Conspiracy to manipulate EUR/USD in the spot market. Traders in an exclusive chatroom, self-described as “The Cartel”, used coded language around the WM/R window. Combined criminal fines of more than $2.5 billion.

  • 2019

    European Commission

    AT.40135 FOREX — Three Way Banana Split and Essex Express

    Decisions of 16 May 2019

    Article 101 infringements in G10 spot: multilateral private chatrooms, recurrent exchange of commercially sensitive information, and occasional coordination of trading, including around WMR and ECB fixes.

  • 2013–23

    S.D.N.Y.

    In re Foreign Exchange Benchmark Rates Antitrust Litigation

    No. 13-cv-7789 (LGS)

    Class claims that banks conspired to fix FX spot spreads and to manipulate WMR and ECB benchmarks. Settlements of about $2.31 billion. A 2022 jury found a spread conspiracy but that Credit Suisse did not participate.

  • 2020–21

    Commercial Court

    Allianz Global Investors GmbH v Barclays Bank plc

    Follow-on and standalone FX claims

    More than 170 investment funds claiming damages for alleged anti-competitive conduct in FX between 2003 and 2013. Pass-on and standing were fought as preliminary issues.

  • 2022–25

    CAT / Court of Appeal / Supreme Court

    Michael O’Higgins FX Class Representative Ltd v Barclays Bank plc

    [2022] CAT 16; [2023] EWCA Civ 876; [2025] UKSC 48

    Follow-on collective proceedings from the Commission’s FOREX decisions. The Supreme Court, on 18 December 2025, restored the CAT’s refusal to certify on an opt-out basis.

Illustrative studies

Questions this specialism is built for.

Typical issues. Not a list of appointments. Barriers and options, 1LOD and 2LOD market abuse in FX, and the rest of the FX book.

FX barriers

  • Chasing a one-touch

    A one-touch digital that paid after a large print in a thin market. Whether that print was a good barrier under the confirmation — source, cut, continuous observation — and whether the tape was a market or a chase.

  • Knock-out of a reverse knock-out

    Whether a print in the observation window was a good barrier under the confirmation: source, cut, continuous or discrete, and what a dealer who had run those books would have treated as a print.

  • Windowed and American barriers

    A barrier live only in a window, or live continuously. Gap risk over a weekend or a data release, and whether a print outside the window was nevertheless treated as a knock.

  • Knock-in never observed

    A knock-in the seller says never traded. Reconstruction of the path, the source the confirmation required, and whether a print was good enough to knock the option in.

  • Double barrier

    A structure with an upper and a lower barrier. Which barrier, if either, was good, and how a dealing room treats a print that tags both in a gap.

  • Barrier in a gap market

    A knock alleged on a print through a gap — G10 after a referendum, EM after a peg. Whether that print was a market a barrier book would have recognised.

  • Source, cut and time zone

    The confirmation names a source, a cut and a city. The print relied on was another page, another time, or another zone. What a dealer would have treated as the barrier print.

  • Discrete observation versus continuous

    A barrier observed at a cut, or continuously. Intraday spikes that would knock a continuous barrier and miss a discrete one — and which the terms actually required.

  • Rebate when the barrier is hit

    A knock-out with a rebate. Whether the rebate was due, at which barrier, and whether the cash amount followed the confirmation or a dealer’s close-out number.

  • Pin risk into a barrier

    Delta-hedging into a barrier as spot pinned. Whether the hedge was managing a genuine knock risk, or a pattern of prints intended to force the barrier.

  • Partial and Parisian barriers

    A barrier that knocks only if spot stays beyond the level for a time, or only on a fraction of notional. Market practice on the clock, and whether it ran.

  • Up-and-out, down-and-out, reverse KO

    Direction of the barrier, reverse structures, and the difference between a regular knock-out and a reverse knock-out in a trending market.

  • Close-out of a residual barrier book

    A book of live barriers torn up when a relationship ended. Replacement cost in the pair, the tenor and the smile — not a vanilla mid from a screen.

Vanilla and exotic options

  • Touch, no-touch and digital payoffs

    A binary or digital FX option where the barrier was close. Determination of the touch, the rebate, and whether the payoff followed the terms or a dealer’s convenience.

  • Double no-touch and range binaries

    A range or double-no-touch that paid, or did not, after a volatile session. Path, source, and whether either barrier was good.

  • TARF, accrual and target-redemption structures

    A target-redemption or accrual forward sold as a hedge. What the payoff actually was, how it accrued, and whether early termination was the market value of the remaining accruals.

  • Collar and participating forward

    A corporate collar or participating forward closed early. Replacement cost of the residual options, and whether the quoted unwind was a market or a penalty.

  • Vanilla close-out off the wrong surface

    A book of vanillas marked to a volatility surface that was not the surface of that day, that size or that delta. What a dealing room would have used to replace the trades.

  • Forward extra and windowed extra

    A forward extra — a forward with a knock-out — where the barrier is the whole point of the structure. Whether the extra knocked, and what remained.

  • Risk-reversal and butterfly marks

    A structured vanilla package marked off a smile that did not exist in that tenor. 25-delta risk-reversal, butterfly, and the surface a dealer would have used.

Market abuse — 1LOD and 2LOD

  • 1LOD review of a 2LOD closure

    An independent 1LOD conclusion from the underlying FX orders and trades, not from the 2LOD assessment. Whether the activity is consistent with legitimate market behaviour or may indicate abuse, and what a senior stakeholder should be told.

  • 2LOD SME on alert quality

    Whether a 2LOD closure met the regulatory and internal standard: rationale, supporting evidence, and the quality of the investigation. Robust challenge where the logic will not bear the close.

  • STOR or market observation

    Whether the pattern met the threshold for a suspicious-transaction report or a market observation to the FCA. The evidence in the FX tape, not the narrative in the alert.

  • Manual surveillance for FX

    Design and operation of manual controls for FX where the automated scenario is silent. Procedures, what to look for, and how a control is presented to a steering committee.

  • Spoofing

    Orders placed and cancelled in a pattern that may have been intended to move a thin FX price. Read from the order book, not from the alert text: size, life of the order, and whether a genuine strategy explains it.

  • Layering

    A stack of orders on one side of the book, withdrawn as the other side traded. Distinguishing a working bid from a layer intended to create a false impression of depth.

  • Wash trading

    Trades that appear to have no change of beneficial owner. Crossing, related accounts, and whether the prints were a market or a device to paint a volume or a price.

  • Ramping and painting the tape

    A sequence of prints into a close, a fixing or a valuation point. Whether the activity was hedging or an attempt to set a price that was not the market.

  • Off-market trading

    A print away from the contemporaneous bid and offer. Size, pair, time of day, and whether a dealer would have treated that price as a market or as a transfer.

  • Insider dealing — macro or name-specific

    Purchases or sales ahead of a data release, a central-bank decision or a corporate announcement. What was public, who was in a position to know, and whether the pattern of orders is consistent with information that was inside.

  • Front-running a fixing order

    Activity in the minutes before a disclosed client FX fixing. Whether the desk was managing the risk, or positioning on the other side of it.

  • WM/R 4pm and other benchmark windows

    Orders clustered in a WMR, ECB or EM fixing window. Methodology of the benchmark, the client instruction, and whether the tape is a genuine fixing or an attempt to move the print.

  • Last look as a conduct question

    Information from a rejected last-look order used to trade on another venue. Market practice on what last look may be used for, and what it may not.

  • Communications against the tape

    Chat, voice and the order book read together. Whether the language is colour, or an instruction that explains a print the tape cannot otherwise explain.

Spot FX

  • Last look on a spot order

    Whether a rejection inside the last-look window was consistent with price-and-credit checking, or with holding the order long enough to trade against it. The tape, the advertised window, and what a competent e-FX desk would have treated as a firm offer.

  • Internalisation versus a displayed price

    A fill taken from the bank’s own book at a price inside the published mid. The question is whether that was a genuine internal match or a price the client could not have obtained in the market of that second.

  • Bid–offer in a thin pair

    An emerging-market or crossing pair where the screen was one-way. Whether the spread charged was the market available in that size, or a penalty dressed as a market.

  • Stop-loss in a gap

    A stop triggered through a gap after a data print or a weekend open. Whether the fill was the first available market, or a price taken from the far side of a book that had already been positioned.

  • TCA against the wrong reference

    A best-execution complaint built on a mid that was not the mid of the venue, the size or the second. Reconstruction of a reference a dealer would actually have used.

Forwards and FX swaps

  • Broken-date forward points

    A close-out on a date that is not a standard tenor. Interpolation of the curve, the points a dealer would have shown, and whether the quoted close was interpolation or a penalty.

  • Historic-rate rollover

    A corporate strip rolled at historic rates. Whether the points, the historic spot and the cash settlement were the market of the roll dates, or a structure that locked in a loss the client could not see.

  • Early termination of a forward strip

    A relationship ended mid-strip. Replacement cost of the remaining dates, in the pair and the size, with the credit of the parties — not a screen mid multiplied by notional.

  • Spot versus points in a swap close-out

    An argument over whether value sat in the spot or in the points. How a dealing room actually splits an FX swap when it is torn up.

  • Cross-currency basis

    A basis swap or a long-dated FX swap where the basis moved. Whether the mark used the basis of that day, and whether the quoted unwind was a market in that tenor.

Non-deliverable forwards

  • NDF fixing after a disruption

    The onshore print named in the confirmation was not published, or was not a market. What fallback a reasonable dealer would have applied, and whether the disruption was real or convenient.

  • Onshore print versus the offshore NDF

    A pair — typically BRL, KRW, INR or TWD — where the onshore fixing and the offshore NDF had parted company. Which rate the confirmation actually required.

  • Wrong fixing source in the confirmation

    The source named was not the source the market used that day, or had been retired. Market practice on the fallback, and what the parties had in fact been settling to.

  • Holiday, cut and publication lag

    A fixing that published late, on a local holiday, or at a cut the confirmation did not name. Whether the settlement rate was the rate a dealer would have treated as good.

Valuation and close-out

  • ISDA 2002 replacement cost

    Close-out of an FX book under the 2002 Master Agreement. Quotes obtained, quotes not obtained, and the difference between a bid for a small ticket and a bid for a residual book.

  • Valuation-agent determination

    A determination under the CSA or the confirmation that one party says is not a market. What a dealer would have shown for that pair, tenor and credit on that day.

  • Illiquid exotic close-out

    A barrier, TARF or accrual book for which there is no screen. How a dealing room actually replaces that risk, and what a mid from a vanilla surface does not capture.

  • Default and a disputed Independent Amount

    Variation margin and independent amount after a default. Whether the close-out amount followed the 2002 methodology, or imported a number from a system that was not the market.

Hedge-fund and proprietary books

  • Unwind of an FX derivatives book

    A $100 million-class FX book taken down when a relationship or a mandate ended. How that book would have been traded, hedged and unwound by someone who had run one.

  • Volatility relative-value versus a barrier book

    A dispute over whether a set of trades was a genuine RV strategy or an unhedged barrier. The Greeks, the tape, and how those books are actually run.

  • Cross-currency correlation book

    Marks and close-out of a book that lives on correlation and FX volatility rather than on a single pair. What a screen in one pair does not tell you.

  • Unauthorised trading in an FX franchise

    Whether a sequence of trades sat inside the mandate of a prop or hedge-fund desk, or outside it. The limits, the strategy, and the tape.

Best execution and dealing practice

  • Agency versus principal

    An order that was described as agency and filled as principal, or the reverse. What the client was shown, what the desk did, and the market available to a client of that standing.

  • Voice versus e-FX versus DMA

    A fill taken by voice when an electronic price was live, or the reverse. Channel, size, and whether the fill was the market of that channel.

  • Algo working a large FX order

    A child-order pattern that may have been a genuine working algo or a pattern that information-leaked the parent. Participation, venues, and last look on the child orders.

  • Prime-brokerage give-up

    A trade given up to a prime broker at a price the executing broker says was the market. Spreads, time-stamps, and whether the give-up price was the execution price.

If the file is one of these questions — or sits next to one — a confidential summary is enough to start a conflicts check.

Instruct