This is extremely complex and it's taking me a few reads to understand it, but I just want to plug this particular journalist as extremely good in reporting complex subjects in a way the rest of us can understand. Do check out his previous article on how T. Rowe Price voted for the Dell buyout by mistake.[1] Truly a great job at reporting.
The article is definitely worth reading and I agree the journalist did a great job simplifying the topic. I also liked this section near the bottom explaining the mindset of the traders:
Different financial markets have, over long periods of time, evolved standards of behavior that are not, perhaps, entirely honest in the most traditional sense of the word. To outsiders -- and to some less experienced participants in those markets -- those standards can seem shocking, even criminal. To regular participants, they can seem normal, even admirable.
And in all fairness if you look closer at any other industry you will find similar grey-ish shady practices.
Planned obsolence in consumer electronics? Intel selling you an i7 for 5 times the price of an i3 when really it is pretty much the same piece of silicon, just with a slightly different design (or some features turned off). Some of the stories I hear in the construction industry are pretty appalling, unsuspecting customers are always ripped off. Pharmaceutic industry? Retiring and repurposing the same molecule to a different treatment to ramp up the price,... Should we talk quality controls in clothing?
Commerce is always inherently a conflict of interests between a buyer and a seller, and what keeps people honnest is competition mostly, and laws that require the seller not to lie about the product or service.
I don't think any of this is specific to financial markets.
think of something like this: you go to a big box store to buy a bunch of home electronics for a new house. you ask the salesperson to throw in some freebies because you're going to be buying a lot. he knocks off the delivery fee, installation fees, and you push further and ask if you can get some other things, say a handheld vacuum cleaner or a stand fan.
he says "Mate this is the best I can give you, I am no longer making commission on this trade". Or if it's the store manager you are speaking to, he says he's not turning a profit with all the discounts already given.
Really pal? You want to repeat that on a recorded line?
Because if this was the credit market, and if your sales guy said you've got a great price because he paid so-and-so for a security he's selling to you and is not getting much in profit, but in reality he actually paid much lower, he can go to jail [1].
But in car sales, home sales, big ticket items like that, you have all these kinds of folks throwing in a white fib here and there, and they don't get locked up.
whatever, I've made my peace with it. there are people who get away with it and people who don't. some people are born rich and some have to work for it.
this hsbc guy is probably going to get into trouble for what he did. the doj is going all out to stamp out this kind of misbehaviour - and not defending it, lying is a scummy move - because the banks themselves could not do it. but while they're at it, maybe start looking at car salesmen and real estate agents too?
WOW. I am really glad you shared this, thank you. I signed up and then my dad was like hey how do I subscribe and I was like uhhhhhhh good question. Glad this was here, thank you.
I remember reading that piece, and now with this one, it's hard to tell if the guy is just that sarcastic, or honestly attempting to defend the practices of the accused. "Listen guys, it sounds bad, but actually this is just how trading is. Why heck, if the trading desk isn't out to bend their customers over, they might even take a loss!"
Not sure why I'd expect any better of a trade rag I suppose.
His basic point is that even though the facts of this case look bad, at the end of the day even normal FX execution will like front running just because of the structure of the market.
why would any legitimate buyer of currency go with someone who might try to frontrun them?
Why isn't there a way for buyers to pay a set amount in fees for the transaction, in order to guarentee that they don't get boned in the back by their very own agents?
> why would any legitimate buyer of currency go with someone who might try to frontrun them?
Because paradoxically it's often cheaper to do it this way than alternate routes of execution.
For example, you can (for a fee) get set up with a trading terminal and just execute against an FX exchange yourself. However, you might end up paying more for that execution that if you went to a bank, especially if you don't know what you're doing.
In theory, the bank knows the market well and can execute your order at a better price (by 'working' your order, splitting up and not tipping off the market), and then passes off some of those savings to the client.
Whether that actually happens in practice depends on the client being savvy enough about the market to not agree to a bad execution strategy (as we saw here) or not agreeing to a bad price.
3. just want a single publicly available price (the WMR fix) to simplify accounting (fx, financial, mgmt, risk and all the disparate ERP systems that a company can be hooked up to)
so the whole market is set up this way.
oh... the e-brokerages like hotspot or whatever would drool buckets at the prospect of global corporates going direct to market rather than through the dealers, but all these companies have decided to cede control to the banks because they don't want to deal with any of the difficulty of fx execution. and so you have this.
I didn't think I would ever follow a specific "traditional" columnist regularly until I noticed one great piece too many from Bloomberg View on HN, with comments drawing my attention to the author always being the same guy.
Today I read him regularly. In fact, it's the only content source I didn't bother adding to my RSS feed, since I remember to open it every morning.
So yeah, passing it on and mentioning: keep an eye on this Matt Levine guy, he is great.
He's the only individual columnist at a mainstream press outlet that I've bothered doing that for, he's that good. (as in, I follow feeds for newspapers, and feeds for individual bloggers, but only Matt as an individual newspaper writer).
Agreed, I'm a big fan of his writing style: Easy to understand, with just the right amount of wit and sarcasm. Makes seemingly boring subjects fun to read!
> Different financial markets have, over long periods of time, evolved standards of behavior that are not, perhaps, entirely honest in the most traditional sense of the word. To outsiders -- and to some less experienced participants in those markets -- those standards can seem shocking, even criminal. To regular participants, they can seem normal, even admirable.
Indeed, when I did my trader exams back in 2007, the regulations treated foreign exchange very differently from other asset classes as far as pre-positioning was concerned, because of the way the FX market traditionally functioned (e.g. No centralised exchanges), although I believe the regulations have changed in the past few years (i.e. after this particular incident).
It's very telling that the UK regulator presumably didn't see any grounds for prosecution or enforcement action. This is a UK bank dealing with a UK client, in the UK. There's no logical reason for a US prosecutor to be involved.
Oh wow, the Department of Justice actually went after someone at HSBC. Are they trying to compensate for a recent report that concludes it let HSBC skate on its criminal money laundering business because it was too big to jail?
And for background on the blatant money laundering they enabled:
>drug dealers would sometimes come to HSBC's Mexican branches and "deposit hundreds of thousands of dollars in cash, in a single day, into a single account, using boxes designed to fit the precise dimensions of the teller windows."
I used to trade FX. With this bank as well, though I didn't know these two guys.
So there's two ways to buy something: agency or principal. If it's agency, you say to a guy "hey, go and buy me some widgets and I'll pay the amount and you'll get a little fee".
If it's principal, you say "hey, make me a price that you want to trade at, and we'll trade, and how you make that profitable for you is your own problem."
What is front running? Generally, it's where you're the agent, but you use that fact to scoop yourself some widgets before you do the trade for the customer. Now naturally, even if you are not literally executing an order for a client, that flow information is useful to you. For instance you might guess all the clients are spooked because of some news, and after two of them call to do the same trade, you guess they'll all do so.
3/4 pm fixing question. Actually seems a reasonable suggestion to do it at the less liquid fix. A lot of derivatives are fixed at 4pm, and there may well be some guy out there with a strong interest in moving it. I recall the price action at 4pm being very odd on many, many occasions.
Anyway, what's confusing to me is why didn't Cairn just get a firm price from HSBC? You can phone a guy and ask for a few billion if you like, people do it all the time. Alternatively, anyone can get an unknown execution price; the banks provide a GUI to everyone who wants one, and you can click the thing until you've got your amount done. The market won't run out of GBP.
What they seem to have done is give a licence to HSBC to trade their order in a way that was bound to create an unequal conflict of interest. The bank can always make it profitable for itself, because they're in control. The customer is bound to be disappointed when they have both an unknown price and price impact against them.
It's possible that since Cairn would be a "real money" customer things are not quite like when I was a customer. Hedge funds are called "leveraged" customers, and the sales people tend to be different, with a different focus. The FCA might also be taking a view that real money guys need better protection than speculators.
I definitely wouldn't have called in to do a single massive trade. I'd have gotten on the GUIs an clicked my way through, with an uncertain outcome, split across various banks and over a few days. It's not as if the profits from selling a business unit will be lost by the currency move between the sale and the cash disbursement to the shareholders.
Agree that you can't lie to the client about what you've gotten up to. That part is undeniably bad.
> I'd have gotten on the GUIs an clicked my way through, with an uncertain outcome, split across various banks and over a few days.
Because you're sophisticated market participant and have sat on a desk. You understand the price impact of showing potential flow to the Street, and how to break up an order in order to hide your intent. Would you expect a random corporate treasurer know how to do that?
Not to mention that there are structural barriers to what you describe -- it costs money to buy Bloomberg terminals and live market feeds, and not every treasury department can afford to do that.
What upsets people is that when traders interact with less sophisticated clients, the latter tend to get fleeced.
I think it's safe to assume that people who do $3.5B transactions on a regular basis are sophisticated. But people who do this once a decade? They're not going to know nearly as much as the experts.
That really seems like a "No True Scotsman" definition of "sophistication". Regardless, it's not the definition the markets use. For the same reason, I qualify to invest in private company stock by dint of my bank account and annual income, not by passing a test.
Sure, there's the legal and regulatory definition that determines who can invest in a hedge fund or open a trading account or sign an ISDA... But! those are just guidelines that ensure that only people who can afford to get fleeced end up at the table.
The more salient definition is that you're sophisticated if you're able to consistently turn a profit (or just avoid getting ripped off).
In a more perfect regulatory world, these folks wouldn't have been able to make the trade until they were properly educated, but we don't live in that world.
As usual, if you can't spot the idiot at the table, it's probably you.
Well, the guy in Cairn's finance department may not know it, but I think there's a lot of overselling coming from the banks. They've probably bought into this idea that the market is a predator that will eat you alive if you don't hire someone who knows what they're doing. I'm sure the finance guys need to look like they know what they're doing to their bosses too, so I suppose it's limited what the options are. Even though the naive path would have worked just as fine as paying HSBC.
>>Not to mention that there are structural...
I wandered past the finance department of a major chip manufacturer when I was working there. They already have Bloomberg terminals. It's not much money for a F500 company. They also don't really need it; what they need is a GUI from a broker that they can sit and click on. That's free and very easy to do.
From my interactions with corporate treasurers, they are basically short a bunch of put options and strongly disincentivized from taking any risk.
Transacting at the fix 'feels' like it's easy to justify to your CFO (hey boss, we got the same rate as the rest of the market, and here's how you can verify that I'm telling the truth), whereas if you trade on screens and screw it up, you'll be out of a job.
The fact that the bank's front running P&L gets pushed into the fix rather than being fully transparent is also probably a feature, not a downside, for the treasurer.
Of course they are disincentivized from taking risk - their job is capital preservation, not capital growth (that's the job of the rest of the business).
Yes, but my point is that there's an information asymmetry here: treasurers are incentivized to look like they're preserving capital, and that doesn't always align with actually saving the company money (e.g. learning to trade off screens, which might be better than just taking the fix from a bank).
You could look at it two different ways: the cynical one is that treasurers are paying out from the company purse to marginally increase the chances of keeping their jobs.
The more charitable one is that doing the 'safe' and 'boring' thing is the most economically efficient way to let the CFO and the board know that the balance sheet is safe, and that's exactly what the treasurer is doing.
How that shakes out in a company's internal politics becomes interesting when you have a case like this one.
Credit checking & per institution limits wouldn't let you trade massive volumes anyway... and even if it did automated hedging splits orders up so as to not signal the market.
The "insider trading pound sterling" claim isn't just less interesting. Whoever put that in the complaint should get Rule 11 sanctions for making frivolous allegations.
TIL: there is no Rule 11 equivalent in the Federal Rules of Criminal Procedure. Interesting discussion at United States v. Aleo, 681 F.3d 290, 308-09 (6th Cir. 2012) (perish the thought that a prosecutor could get sanctioned for filing a frivolous pleading that would get a civil litigator Rule 11'ed).
The author alludes that the traders would have been making those trades "all day", but suspiciously absent is any evidence that they actually had. According to the facts reported, the traders only traded pounds on three days over a week-and-a-half period.
We need to see the traders' trades for a longer period before the Cairn trade to see if they actually do trade pounds all day.
The person arrested was "global head of foreign exchange cash trading". Is it normal for global heads to be making routine trades all day? (the article indicates it is not; again, only three trade days out of 10+)
From the article it sounds like this person was more involved in client relations/sales than technical execution.
By the same token though, the entire FX book is under his jurisdiction, so it wouldn't have mattered if he'd delegated to another trader (except that it looks more culpable this way).
"HSBC gained approximately $5,000,000 from its execution of the Victim Company FX Transactions," or about 0.14 cents per dollar of the $3.5 billion transaction."
Am I misunderstanding what this says or is the amount of profit about two orders of magnitude off?
$5,000,000 profit / $3,500,000,000 transaction volume = $0.0014 profit per dollar transaction volume. 0.0014 dollar are 0.14 cents. Seems correct to me.
Overall, I appreciate the article, but it seems to oversell the claim that this wasn't fraud. They made a purchase on behalf of a client, then lied to that client about when they were trading and who else was trading.
The front-running claims are messy at best, but there are other claims here that seem to be bulletproof. When someone hires you for a transaction, and you explicitly, provably lie to them about the nature of that transaction, surely that's fraud? HSBC made claims to a client that were blatantly false, expressly for the purpose of extracting more funds from that client's transaction.
That's the thing though: Technically and legally, they did not do this.
An agent buys things on your behalf from the people who are selling them; they have an obligation to obtain the best price for you. A principal is selling you things which they bought on their behalf; they have an obligation to sell to you at the price you agreed. And in this case, HSBC was a principal, not an agent. They made no purchases on behalf of the client; they made purchases on their own behalf so that HSBC could fulfill the order that the client had placed with HSBC.
It's awkward though because it looks a bit like they were buying on behalf of a client, and it probably felt to the client like they were acting as an agent, and the price they agreed on was one that HSBC had some control over, which is really not a good idea at all without an agent relationship. And yet: The purchase was not being made on behalf of the client and HSBC was not an agent of the customer.
> surely that's fraud?
Simplifying a lot: Only if the customer relied on those lies. If I sell you a used car and I say the last owner's first name was Bob, and it was actually Bill, and you later regret the purchase, it's not fraud, because you can't truthfully say that but for my lies you would not have purchased the car. If I say the last owner was a little old lady who drove it down to the shops once a week, but they were actually an uber driver who drove it 60 hours a week, then maybe that was a material and would have impacted your decision. (That's a question for a jury.)
Offhand, the lies about Russians sound horrible, but I'm not sure what impact it had on the customer's decisions, especially since it happened after they agreed to the trade. Now, the lies about liquidity to get them to agree to the 3pm fix might qualify though...
One problem with chasing after these fraud claims is that it's difficult to demonstrate, after the fact, how much the client was harmed by the fraudulent statements.
For example, would the client have traded at 3pm instead of 4pm without the misrepresentation about liquidity? Obviously, the client will claim now that they would not have, but they're not an unbiased party.
1] http://www.bloomberg.com/view/articles/2016-05-13/t-rowe-pri...