Hacker News new | past | comments | ask | show | jobs | submit login
Big investors sue banks in U.S. over currency market rigging (reuters.com)
162 points by petethomas 5 months ago | hide | past | web | favorite | 56 comments

Former FX fund manager here. I believe the allegations.

What happens is that most derivatives like options or variance swaps are tied to the daily WM/Reuters fixing in the London afternoon. There's a bunch of different types of derivatives, but the thing about most of them is they have some sort of characteristic where the price dependency gets highly nonlinear towards an expiry.

So this means that someone who had a moderately sensitive position on one day might have an extremely sensitive position a few days later.

To the point where it might make sense to make sure the price doesn't hit a certain level at the fixing, by doing a bunch of trades in the spot market leading up to the fix. Or conversely by making sure it does hit some level by ramping up the price.

Now you might think it's all more or less a wash, because someone is gonna have the other side of that derivative, but that's not the case. The banks tend to have the same sides of the trades against their customers, because the customers are mostly all after certain payoff schemes.

Apart from there being a motivation, I also believe the allegations because I've been told the actual positions on certain days. You'd have these days where nothing was happening at all, and then just before the fix the price would ramp, or the price would be moving but the graph would have a weird flat ceiling. So you'd call the brokers and ask him WTF happened, and they'd say something like "XYZ bank has a huge barrier there", or "Bank X has wants to knock out this level".

It's very apparent to anyone who looks at any product that has a fixing, I've traded several (Swaps, FX, Equity Derivs). You start by thinking it's just noise and there's always someone in the office who will say that, but after a while you get suspicious of it happening at the same time, plus you have the broker rumours lining up. It would be great if the free market were restored.

Similar case in the rates world is LIBOR fixing. It's an outdated scheme that managed to escape regulatory attention for a very long time. By design there is a trust issue yet so much of the global financial system depended upon it. I have left finance years ago, but there was a movement to move away from LIBOR as funding index. I'm not sure if that ever happened.

LIBOR is being replaced by SOFR and its siblings. They're starting to get reasonable traction.

Fwiw, from an outsider's perspective, I hear more and more grumbling about LIBOR, lately. Disilliusionment with it is reaching the main stream, for better or for worse.

LIBOR met the mainstream decades ago, in the 80s and 90s a nice guideline, and ended up pre-financial crises with adoption of financial models that balanced on the edge of single basis points that depended on a single digit that a not very well paid team calling traders, that might or might not answer their phones, and might or might not care to give an accurate number, but when a number it was not to single basis-point accuracy, just a number from the floor.

The people playing with complex quantitative systems that depended on single basis points had no idea of the integrity of the data they were playing with if they were using LIBOR.

The conclusion I got was: Get out of your code and check your data and how much you trust it. Then get back in with a whole load of caveats.

> It would be great if the free market were restored.

This is what happens in a free market.

Exactly. A perfect market requires these things:

- Perfect market information

- No participant with market power to set prices

- Non intervention by governments

- No barriers to entry or exit

- Equal access to factors of production

- Profit maximization

- No externalities

The less you have those, the less efficient your market is. "Free markets" aren't these magical things that arise on their own in nature. They are carefully crafted and maintained artificial environments and most are far from efficient.

Market forces are not a silver bullet that will solve all problems if we just sit around and wait long enough for the invisible hand to do everything.

Many would include trust, non-collusion and other acts to be against the concept of free market/trade.

Although some would draw the line at different locations, but anti-trust, collusion, patents etc. part of the system. It's just a matter of how much to limit either the government and public good vs. the corporations.

In this case, collusion in trade is obviously a violation of anti-trust law.

"Now you might think it's all more or less a wash, because someone is gonna have the other side of that derivative, but that's not the case. The banks tend to have the same sides of the trades against their customers, because the customers are mostly all after certain payoff schemes."

Thanks for clarifying. That was really confusing me because it did sound like it ought to be a wash.

The other thing is: is this illegal? I am assuming there are no insider trading laws that apply to FX because there is no fiduciary duty to a currency.

>You start by thinking it's just noise and there's always someone in the office who will say that, but after a while you get suspicious of it happening at the same time

I felt the same way when I used to watch bitcoin. It's the coherency that is oddly convenient.

Hey could you recommend some books for me to read to better understand what you’re talking about and have a similar level of comprehension of financial matters such as this


> "Former FX fund manager here. I believe the allegations."

I believe the courts agree with you. That is, per the second paragraph, this is not new news, simply a new lawsuit (by entites that opted out of the previous suit).

I'm curious to know what sort of timeframes they were manipulating on. Most if the large institutional banks are operating as market makers, which would likely limit their influence on price movements to a few pips at most. Most institutional investment firms in forex are not trading as market makers though...typically operating on much larger time horizons...hours to days. The amount of capital necessary to manipulate the currency on those timeframes is incomprehensible. I doubt even the 16 largest banks in the world, even working as a deal facto cartel, could coordinate price manipulations bigger than 50 pips over the course of a day. I mean we're talking about a quarter of the GDP of the entire US changing hands every single day via these markets. These banks are massive, but still nowhere near big enough to materially affect the markets. It makes me wonder what sort of damages are being claimed, and how they are calculating those damages.

I am not close to the matter but my understanding is that they were not accused of pushing prices throughout the day, just around the time of a key fixing that would be used as reference for the payoff of derivatives.

If that's the case then it seems an easy solution (if the parties involved are interested in a solution) is to not let any reference point be so brief, but instead an average over a whole day or even two days.

no, not 50 pips, most likely just enough to get their options contracts to expire at the right place if some junior trader was managing a position around a certain strike

pip = percentage in point, smallest amount a particular exchange rate can change, see https://www.investopedia.com/terms/p/pip.asp

Thanks !

Puts all the "cryptocurrency is a scam/fraud" into perspective.

Does anyone here know the actual mechanics of what's being alleged? In what sense were these individuals 'manipulating' these markets? The only method I know of to do that is to spoof orders...but they would have to be making astronomically sized spoofs to move the FX markets.

This all came out some years ago; this is just a followup lawsuit from the people who opted out of the first round of litigation.

In short: It didn't work very well, but that's not the point.

Longer form: Traders from several banks shared client positions, and made very enthusiastic if somewhat amateurish efforts to manipulate the price at the close. The core of it was trying to figure out if there was going to be a lot of pressure one way or another on various price pairs, and then trying to manipulate the spot price at the close to take advantage of it. For example, if you know a few large clients will be selling a net of £100m for USD, you might try accumulating a bunch of USD slowly in advance of the close (to avoid driving the price up), so when the sales hit you can profit. Or you might try and buy a little USD early, then hold off until right before the close, and buy a bunch very quickly to try and spike the price heavily, making your earlier purchases more valuable. Etc.

None of it sounded especially like it would work, many of their techniques were completely contradictory, and nobody really bothered to try and calculate if it did work. The FCA said "that it is not practicable to quantify the financial benefit". One example of a very good trade apparently made US$100k for the offending banks (on a volume of over $500m); many others lost money. Total profits from the activity were at most a few millions of dollars per bank per year, if they even came out ahead; their fines were orders of magnitude higher.

Or as the CTFC described it:

> If traders in the chat room had net orders in the same direction as what they desired rate movement at the fix to be, then the traders would at times either (1) match off these orders with traders outside of the chat room in an attempt to reduce the volume of orders in the opposite direction transacted during the fix period; (2) transfer their orders to a single trader within the chat room who could then execute a single order during the fix period; or (3) transact with traders outside of the chat room to increase the volume traded by chat room members during the fix window in the direction favored by the private chat room traders

In other words, to try and drive the price of GBP up the traders might 1) try and buy GBP from outside banks to stop those banks from selling GBP at the close 2) not do anything in particular 3) try and sell GBP to outside banks so that the traders wouldn't need to sell GBP at the close. It seems pretty unlikely that these are all valid strategies, and indeed, they mostly didn't seem to work.

> In what sense were these individuals 'manipulating' these markets?

Primarily in the sense they had intent. The actual impact seems questionable. But with chat transcripts that damning, what more do you need?

(Matt Levine has written about this extensively, among other places, here: https://www.bloomberg.com/opinion/articles/2014-11-12/banks-...)

> "This manipulation was allegedly done through chat rooms with such names as “The Cartel,” “The Mafia” and “The Bandits’ Club,”

If this is true, it would seem most unwise to discuss your cartel in a chatroom named "The Cartel".

How high-level would employees in these alleged rooms be likely to rank at?

Former finance employee here.

We had a blanket ban on any project or program name that was even slightly insidious sounding, which is easy to do on accident when there are hundreds of projects and the name will be examined out of context by a regulator with 0 sense of humor.

While it’s possible that the “cartel” started with that intention, it’s also likely that someone thought it sounded cool at first.

I once worked with a quant that had a batch of strategies he was testing. The one that was performing the best he named ‘the front runner’

He was shocked when the firm required him to rename it & not use that name in any communication.

> which is easy to do on accident

One of the first things I did when I took over a trading team was turn off chat. Trading involves stressful, adversarial interactions with counterparties and clients. Every person and deal eventually earns a nickname. It helps nobody when those nicknames appear in writing.

Wise choice.

  it would seem most unwise to discuss your
  cartel in a chatroom named "The Cartel".
In the early 2000s, a group of corrupt Alaskan politicians literally had hats printed saying Corrupt Bastards Club [0]

Some people have odd ideas about Linguistic reclamation.

[0] https://en.wikipedia.org/wiki/Alaska_political_corruption_pr...

It's like using "password" as your password. The people who do that think they're being original and clever.

That's silly. It is only original and clever if you replace the s with $ and a with @

Associate level, or really, really dumb Vice Presidents

In businesses that deal with economics everybody is an associate or a vice president

Except for Analysts, Managing Directors, and those at the Partner level. But OK, if that's what you'd like to believe.

Worth reading for those concerned with rigging in much smaller crypto markets

What haven't these banks rigged. It is not limited to the US. In the UK Barclays Bank have rigged LIBOR and in Australia the biggest banks ANZ, NAB and Wespac also rigged the bank swap rate.

What this means is that we all pay higher interest rates on our loans, and the bankers get a tiny fine.

Australian regulators waxed lyrical about how "it's different here" at the time, both sides of politics said the same.

3 years later and the banks ended up in court for interest rate rigging, all the regulators decreeing innocence from upon high now have lucrative private sector finance jobs and the relevant ministers lying through their teeth have long moved on.

No one held at all accountable for supporting a rotten industry, in fact the opposite. Banks are the new untouchables.

> What this means is that we all pay higher interest rates on our loans

Manipulation is manipulation. Lots of circa 2008 rates manipulation pushed rates down, not up. Ripping off investors versus borrowers isn’t an excuse.

Except that the manipulations that banks were accused of resulted in lower libor fixing, i.e. you paying less on your loan than you should have. Investors are the one who can complain that they were underpaid.

The technical problem is that both Libor and FX have a batch closing cycle…how quaintly Victorian.

The “close” (aka the “fix”) has a real need. Lots of financial systems need to enter a FX or Libor rate. You cannot enter a real time rate. You need a rate on a daily basis. I am sure all systems will go real time at some point, but that does not solve the problem today.

It is technically simple to come up with some average for the day for both Libor and every FX pair. I can imagine different algos for this, all open source (eg. trade weighted average). Systems that need a daily FX or Libor rate can choose which algo they use. If one algo gets gamed too much, you can switch to a different one.

That would be a simple surgical fix. It would move the last vestige of Victorianism from our global financial markets. In the meantime, I am sure we will get a lot more pain-killers and bandages in the form of lawyers listening to data mined chat rooms (which will simply force the bad guys to collude in the sauna or other place where nobody can listen).

What confuses me the most is that with all these issues (LIBOR, FX, australia etc) there is no open independent exchange used - ala a stock exchange. Instead there are various OTC (read private/not public retail) - with weird discovery mechanisms like ringing people and asking what rate would you lend me million quid at (Yes Libor was a series of phone calls!!!)

I cannot prove it (would be interesting) but most financial scandals - and probably most risk - would vanish if most trading was open independent exchanges - am I missing something?

> am I missing something?

Yes and no. Firstly, these sort of manipulations happen on exchanges too. Historically, the main difference has been that it's in the exchange's interest to spot this stuff quickly and stamp it out.

Also, and this is asset dependent, it's generally easier for an individual to manipulate a stock price than FX or Libor (back in the day) as they're much less liquid. As such, more focus was put on detecting and prevention.

For many FX markets, it's practically unfeasible to manipulate the market without collusion. This is because of the huge liquidity. So the reason why this was a big deal, and the reason for the convictions, was because of the collusion. Without that there wasn't much to see here (arguably there wasn't much of an effect anyway but the private prosecutions should throw more light on that).

Libor is an interesting case. Similar to FX, it was nigh on impossible to manipulate in a liquid market without collusion. It is true that the rules do not apply to an observable rate but, in practice, it was trivial to tell if a single institute was doing something fishy so they didn't.

Where it all went to crap was when the banks stopped trusting each other. Libor is the uncollateralised borrow rate. In the period preceding the crash everyone stopped lending uncollateralised as they weren't sure they would get their money back. As such, the premise of Libor became flawed as it depended on an answer to the question "what rate can you borrow uncollateralised?". It didn't have an option for "there isn't one", and worse, the setters were pressured into not giving an indication that that was true.

So, in the Libor case, if the borrowing had been on an exchange and the rate set from an observable price, things would have been better in the sense that it would be obvious that there was no liquidity (although everyone knew that anyway). But, ironically, it would likely have been much easier to force material moves as a single entity because of that illiquidity. And we'd still end up with the same convictions as they were due to collusion.

Finally, for FX, it's not obvious that an exchange would help for the same reasons. It's difficult to manipulate without collusion because of the liquidity. An exchange won't change that. And exchanges are no less susceptible to collusion than the current FX market. Arguably, they may be more so if there were multiple exchanges each with less liquidity but that's just speculation.

> I cannot prove it (would be interesting) but most financial scandals - and probably most risk - would vanish if most trading was open independent exchanges - am I missing something?

Open independent exchanges can be and are being rigged and abused too.

A little background:

First, in the FX market it's quite common to make trades "at the close", ie, the price at a specific time. In principle this serves everyone's interests, because you need to agree to some price, and "the price it happens to be at 4pm" makes about as much sense as anything, and it means that customers can easily check to see what the price was at that time, confirm they paid it, and feel good that, while they may not have got the best price going that day, at least they got the "standard" price.

(It might be tempting to tell your bank you want the "best price that day", but obviously, you only know what that price is after the fact. You could hire the bank's traders to take a wild stab at guessing when it might be, but people who can reliably predict the FX markets are too busy sunning themselves on their yacht made out of gold plated diamonds. Better to just take the price at the close.)

Second, the bank, having agreed to sell your pile of GBP for USD (or whatever) for whatever the price at the close is now have some risk. They've got to go and buy all that USD, and then once they swap it with you for your GBP, they have to then go sell it for USD, so they can sell that and get back to where they started. And what if the price crashes in the meantime? So as is proper (not to mention generally legally required), banks hedge that risk, by, eg, selling GBP and buying USD in the run up to the close, to make sure they're covered if prices move against them.

(Hey, doesn't all that hedging actually make GBP cheaper and USD more expensive? Meaning you'll get less USD for your GBP at the close? Why yes, it does. If you tell your bank to sell £50m for you for whatever the price is at 4pm, you would expect to see them busily driving the price down in the runup to 4pm. That'd just good hedging, and it's perfectly legal.)

Third: What's not legal is making a chatroom called "The Mafia" with traders from other banks, sharing client information, and talking about "taking out the filth" or "front running". Again, it's fine if it's just within your own bank, and you properly disclose it in your fine print, and you don't call your customers names; the issue here is more of branding that substance. There's nothing wrong with knowing an insurance company is about to sell you £50m, which means you'll be selling £50m so you frantically run around selling it in adance...as long as you can plausibly claim to be doing it hedge your trading risks. If you're doing it a chat room called "trading risk compliance committee", that might be plausible. If you're doing in a chat room called "The Mafia", even your lawyer will struggle to keep a straight face.

Fourth: The banks in question already got sued and had to settle for $2.3b in penalties; it was in discovery that all the colourful chat room names came out. So it's not in question whether this happened; it did. Nor has any new information came out (the stuff about the chat rooms came out years ago at this point); this is just the residual lawsuits from the people who opted out of the first one.

In short: This is interesting primarily in a horrified "I can't believe these idiots said this stuff where it was being recorded" sort of way, not because what was going on was actually that bad. It's not even clear that the traders in question actually made money from it, or that any actual customers were harmed. But it seems clear they intended to make money and harm customers (on purpose, that is, because again, you're allowed, and in fact, strongly encouraged, to hedge your trading risks, including the risk that your going to lose a pile of money from your agreement to sell a bunch of GBP on behalf of a client this afternoon, and the way you hedge that risk is to sell it now.), so...billions in fines. Good job guys.

> Deutsche Bank

As a German I have to ask what is wrong with this sick piece of a company. They seem to have their ugly fingers sticking in every pile of dirt potentially hiding gold nuggets. Disgusting.

We say the same thing about Goldman Sachs. I guess every country has their own villans.

Massive breach of trust but what can we expect when such entities are private with little to no oversight

I'll argue that such things happen regardless of the presence of the oversight. American financial markets are surely quite bureaucratised, but the fact that such things still happens in countries that beat US to pulp when it comes to bureaucratisation of financial industry puts weight to this argument.

I'd say that the biggest aggravating factor to this is that financial industry is not competitive, not only in US, but pretty much all around the globe. Huge regulatory pressure surely hampers competition.

> Huge regulatory pressure surely hampers competition.

Recently decided to open a bank account with another bank so I can shop around for time deposit rates. The form is 29 pages, almost all to satisfy regulatory requirements.

The price of one of my medications just tripled (to over $1000/year) due to the limited number of government approved suppliers, one of which bought the products from the other and then shut the cheaper competing product down.

> We're from the government and we're here to help ... the industry who captured us

> The price of one of my medications just tripled (to over $1000/year) due to the limited number of government approved suppliers, one of which bought the products from the other and then shut the cheaper competing product down.

Maybe read up on countries with proper healthcare systems. The problem you describe is very much a US problem. I'm very happy with the healthcare system of my over regulated and bureaucratic country (the Netherlands): everyone pays insurance starting from less than a $100 a month (but if you don't have enough income the Government will give you the money to pay for it) and as such everyone receives healthcare without crazy bills (assuming you take medication for a medical condition).

Most people won't believe that it only costs $100 a month in health insurance for you! Incredible. Is the coverage good (pre existing etc)? Or are there nasty surprises (buried in a 5 point font somewhere) that you only come to know when you fall sick?

The only extra cost one might have is "eigen risico" (self risk). Meaning that the first time something happens in a year you need to pay the first x euro. But I don't think that applies to everything. I have never paid that (but I rarely need medical help).

As a more practical example: I travel to Asia a lot and as such I wanted to get some shots for tropical viruses we don't have in our country (yellow fever, rabies, etc - note that we do think vaccinations are good). I got around 10 shots in total, I think the bill was ~1200 euro (~$1500). I forwarded the bill to my insurance company and they paid it back 100% no questions asked (note that the shots I got were not mandatory and I wasn't sick, just wanted to prepare for traveling).

It's a kingdom


Big banks almost always come across as criminal enterprise. They have years of heads up and legal/compliance red tape figured out. Bitcoin has it’s own centralization problems but in the long run (100 years), it can be a better alternative

Bitcoin does nothing to prevent this from happening. It’s a question of how two currencies are priced relative to each other, but you could get the same effect with oil or other commodity.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact