There is nothing so complicated that, with their talent for ingenuity, humans cannot make even more complicated. This applies to IT as much as it applies to finance.
Trading used to be limited by how fast one human could shout at another and agree upon a price. Now it's limited by the speed of an electron through copper wire. This has caused, to put it mildly, some changes. In April 2013 bombs went off at the White House and Barack Obama was injured, the Associated Press reported. The news …
Friday 21st June 2013 10:28 GMT Tanuki
Friday 21st June 2013 11:10 GMT C 18
Listen child, electricity, what is that stuff being stuffed down the copper wires travels at the speed of light. Some people even think light speed in a solid is slightly faster than the speed of light in a vacuum, but there's a lot of faffing about in that debate.
Regardless, electric signals move at the speed of light. And not in tubes, they're always running late so don't serve as a good metaphor for speed. But the Tube is a good metaphor for a signalled network...
Hmm...anyway, try to keep the noise down. We're trying to sleep, perchance to dream for what stuff they're made of or something like that.
Friday 21st June 2013 11:54 GMT Anonymous Coward
Re: Oh dear...
If you're going to be pedantic, it is your responsability to supply correct information.
The speed of the electromagnetic wave is approximately the speed of light in a vacuum. In air, it's not mucj slower. In wires however, it is basically determined by the dielectric constant of the conductor and to some extent of the insulator.
It has, for example, been determined that EM wave speed in coax is in the 65-70% of LS range.
The electron speed, however, is closer to walking speed. In fact, the electrons are more akin to bungling drunk footie supporters than the Millenium Falcon
Have a nice weekend.
Friday 21st June 2013 12:20 GMT Rosie Davies
Re: Oh dear...
Actually dear, electricity (or at least electrons) more very slowly through a copper wire (http://en.wikipedia.org/wiki/Drift_velocity).
The reason things seem to go so quickly is that for one electron to hop across atoms all the others must get out the way and you end up with something sort of resembling a really tiny domino toppling session.
Friday 21st June 2013 13:51 GMT Tanuki
Re: Oh dear...
I think you need to do some study of issues like "Velocity Factor" and "Dielectric" before you assert such things.
[I've got rather a lot of experience of this sort of thing and if you can actually get a signal to propagate down a piece of copper wire at the speed of light I can arrange you a cushy job in the phased-array-antenna design department of my current gig].
Friday 21st June 2013 10:44 GMT Anonymous Coward
Performance is everything
I once had a job interview for a test position within a company which provides trading software to one of the world's largest markets.
What surprised (and frankly scared!) me was that, talking to them about usual test planning, automation etc, that they weren't overly concerned about the correctness of the data, but purely focused on the speed.
I guess a few trading packets dropped here and there are nothing compared to selling up shares in DoomedCorp instantly....
Friday 21st June 2013 10:48 GMT Anonymous Custard
Friday 21st June 2013 11:14 GMT lnLog
Friday 21st June 2013 14:13 GMT Anonymous Coward
I disagree for many reasons. The first being the obvious one: people pay to be able to trade on the stock market. So where the article says that in some transactions millions were lost, the other side of such situations is that there were also plenty of people who made lots of money out of it. Its the beauty of the stock exchange in my opinion: almost everything works both ways.
Another aspect is the damage itself. What damage exactly? Trading or investing on the stock exchange involves taking certain risks. You can try to regulate all you want, but its those risks which allow people to either make or lose lots of money. Regulating only means postponing the inevitable.
And that's problem number two: If you regulate the amount of trades people can make you're not only blocking possible losses, but also blocking possible profits. Because having the option to purchase many stocks in one go is a good way for a trader / customer to invest and for the company which sells those stock to make some profit out of this.
And it's those profits why most people trade on the stock exchange in the first place.
Friday 21st June 2013 18:18 GMT Anonymous Coward
@ShelLuser - Re: @InLog
Let's put a (progressive) tax on transactions then and let those HF traders regulate themselves. If they are rich enough to spend money on these gadgets, they should have enough for the taxman too. Taxes are the best way to put a brake on something without using any interdiction.
Friday 21st June 2013 19:09 GMT Intractable Potsherd
Monday 24th June 2013 10:18 GMT Squander Two
"However, these greedy, panicky fuckwits affect *my* life"
Except not. The Parliamentary report into the '08 crash is out -- and hardly a whitewash, since Parliament have been at the front of the queue to complain about evil gambling investment bankers -- and they've concluded that Northern Rock went bust by lending too much money, mainly mortgages, to people who couldn't afford to pay it back, HBOS went bust by lending too much money, mainly mortgages, to people who couldn't afford to pay it back, and RBS went bust by paying too much for ABN Amro and lending too much money, mainly mortgages, to people who couldn't afford to pay it back. It was a retail bank fuck-up; not an investment bank fuck-up. HFT, complex derivatives, naked short selling, and credit default swaps had nowt to do with it -- and this is coming from the people who assured us that the problem was entirely caused by HFT, complex derivatives, naked short selling, and credit default swaps.
Monday 24th June 2013 10:50 GMT Professor Clifton Shallot
Is there a benefit?
This really isn't my area of expertise so I hope someone on the Reg can help me become better informed - is there any benefit to all this rapid trading? It seems to me that only the winning participants benefit and that there's no way for me to reliably predict who they will be.
So would I care if, say, each traded entity had a price set daily at which trades could be made and which did not change until the next day?
Monday 24th June 2013 12:56 GMT Billa Bong
Re: Is there a benefit?
I can't tell whether this is a sarcastic comment or not... The point is to _be_ the fastest so you can _be_ the winning participant. If a company announces lay-off's, it can't be doing very well, so the trick is to catch the news and put in an offer to sell all your shares at a (not so reasonable) price before everyone interested in buying realize that you're now trying to offload something that's not worth what you're asking. I.e. the trick is to offload something worth 2p for £2 before everyone else realizes it's not worth £2. At the end of the day rapid selling only helps you increase the gap between what "everyone else" thinks a stock is worth and what you think it's worth.
Now scale that up - not only do you need to watch the news, but you also watch what other traders are doing, trying to infer meaning from their activity. The big players can really screw up the market simply by deciding to cash in and sell all their shares for _no_ reason, resulting in other algo's figuring that "they must know something we don't" and also selling up - no human intervention, yet the value of the stock falls significantly.
The principle of a free market is that I can buy whatever I want provided I'm willing to pay the price being asked by the _seller_. If you as the exchange enforce a fixed price, you're effectively saying "yes, you own it, but I'm telling you that you can't sell it for more/less than this price". It doesn't make a whole lot of sense. I'm sure you wouldn't like it if ebay told you that your car is only worth £6000 today and therefore that's all you can sell it for.
If this was the case I'd also be out of a job since I work for one of the low-latency market data vendors!
Tuesday 25th June 2013 08:39 GMT Squander Two
Re: Is there a benefit?
OK, here's an example. Let's say it's about midday, and the news breaks that Company A is not doing very well -- some layoffs, bad quarterly results, the CEO resigns, whatever. You have a million shares in A, which cost you £1 each and which were worth £1.50 at 09:00. Now, in the light of the news, no-one's going to pay £1.50 for them, but, if the news isn't too bad, and if a buyer reckons the price might rebound, someone might pay £1.10 for them. You might well want to sell at this point; you'd at least have the choice.
Under your suggested system, you're forcing everyone to wait until the next day before buying or selling, as sales simply would not happen at the pre-news price -- you wouldn't spend £1.5M on something worth maybe £1.1M, maybe less, would you? What you're actually doing there is introducing uncertainty and risk into the system: you're denying people the option of offloading risky shares, forcing them to hold on to assets which may be about to plummet in value.
One of the advantages of a stock exchange is that it does in fact set price fluctuation limits. When you hear on the news that some stock chashed and trading in it was suspended, that means that the price dropped so far that it breached the exchange's limits and the exchange therefore disallowed any further trading in that stock. So exchanges allow the price fluctuation that a free market needs while also imposing some reasonable limits on the more extreme fluctuations to give the system a bit of extra stability.
So the benefit for traders is simply that they are free to buy and sell stuff at the price they want to, same as you are, which helps them to avoid risk and catastrophic losses. As for the benefit for the world at large, remember that this entity with a million shares could be your pension fund, or a company with ten thousand employees.
Sunday 23rd June 2013 10:45 GMT John Savard
The important thing is that jobs get lost when the stock market crashes.
The stock market should be regulated, so that it performs its intended function - making it possible to raise money to build productive things like factories - without other things happening that detract from the successful accomplishment of that function.
Letting stock market speculators make profits off the backs of working people is not an important function of the stock market, so if regulations block that it's no loss.
Friday 21st June 2013 18:49 GMT Ken Hagan
Or time to stop bailing out the losers. If you made it perfectly clear that there is no happy ending when your clever algorithm bets the entire bank on the 2:30 at Ascot, this would all be self-regulating (at least in the next generation).
Also, stock markets that didn't allow selected customers to buy themselves into a privileged trading position might suddenly find that sane companies wanted to be listed there, rather than live with an ever-present threat of waking up one morning to find the (real) company has been wiped out has been wiped out by some cheeky rocket scientist on $1m/year.
Friday 21st June 2013 11:21 GMT hammarbtyp
Friday 21st June 2013 12:23 GMT Anonymous Coward
Re: Fascinating article
I work for such a company and whilst it pays well and the work is interesting with new problems to solve all the time in the arms race that is faster execution and better quoting (for us, not you) you cannot help but think "the human race would be far better served if we all stopped fucking around with this bollocks and did something useful". Like others we have physicists (nuclear variety), an actual rocket scientist and numerous maths PhDs. Surely there's a better use for that? However whilst inflation remains uncontrolled and house prices are constantly pumped we all have mouths to feed and roofs to provide thus this will continue.
Friday 21st June 2013 12:49 GMT hammarbtyp
Re: Fascinating article
"However whilst inflation remains uncontrolled and house prices are constantly pumped we all have mouths to feed and roofs to provide thus this will continue."
and do you think there is no a correlation that and financial instruments that no one has any control?
The thing that annoys people outside the financial industry is that all the effort only ends up transferring money from one place to another. It does not produce wealth only redistributes it among those who don't really need it. It is akin to betting where you get the horse results 5 minutes before the rest of the suckers.
If half the effort was put into manufacturing real things then maybe the country would not be the financial mess it is now.
Sorry <Rant> over
Sunday 23rd June 2013 13:41 GMT Mark 65
Re: Fascinating article
"and do you think there is no a correlation that and financial instruments that no one has any control?"
Please, inflation and house prices have everything to do with the Government - do you honestly believe that if they didn't want the price pumping to occur that they couldn't have stopped it? House prices are pumped by the Government for two reasons - taxation, and wealth affect. Housing is the one area where tax is near impossible to avoid. Stamp duty on transfer, council tax on value, and debt slavery for good measure to keep you under control. The wealth effect is just there to convince you you've all got more money in the pot whilst you are systematically getting shafted by the wealth destruction that is inflation. A little sugar coating if you will. Ever notice how the only people with consistently inflation busting pay rises and pensions are politicians? Sure they'll slow it down for a year or two (and suckle harder on the expenses teat to compensate) to keep the voters appeased but they'll always be well ahead of the curve. Public sector wage inflation during the same period was vote buying by Labour as they pissed the country's wealth away. Benefits rose mightily as well to buy a few more votes. Notice the uproar when you try and lower public sector pay or benefits despite the private sector being hammered? Sense of entitlement anyone?
You can ride along on the banker bandwagon with all the other general public suckers that believe the political spin of "it's all the nasty banker's fault" if you like but I prefer not to live in ignorance. The people in the limelight before the financial world imploded were the expense cheating leeches in the corridors of power. The people that pissed away the money in the good times so that we now have a structural deficit in the bad. You might want to ask yourself whose responsibility it was to regulate your favourite bogeyman? Didn't occur to you that what was happening was fine and dandy by them until it went boom?
Take away every single penny of the nasty bankers debt and the UK still has a massive structural deficit. Know what that means? I'll spell it out for you - we're spending way more than we have, every bloody year. Get rid of the city (err, and the billions in taxes generated incidently), manufacture your heart out all you like but the UK cannot afford the payments it is making whether that be for middle eastern wars, benefits, or the NHS.
Get used to it because the country is long-term f*cked.
Monday 24th June 2013 10:41 GMT Squander Two
Re: Fascinating article
> The thing that annoys people outside the financial industry is that all the effort only ends up transferring money from one place to another. It does not produce wealth only redistributes it among those who don't really need it.
Yes, this does annoy people. It's not actually true, but it does annoy people.
Every day, the benefits of an efficient market are reflected in the price and availability of everything you buy. It's often quite difficult to explain the link between exchange trading and how well stocked your local supermarket is and why the country never runs out of eggs or lightbulbs, which is why so many people are unaware of its existence, but the link is there. Look at countries that aren't properly involved in exchange trading. Compare the difficulties faced by their populations with ours.
The pricing mechanism creates optimal prices, which in turn maximise trade. By exploiting market discrepancies, arbitrage removes those market discrepancies, giving us more optimal prices sooner. HFT, by speeding up arbitrage, gives us optimal prices even more quickly. The more optimal the prices, the more efficient the entire system by which we produce goods, move them around, and sell them to people who want them. What would happen if some disaster led to the wholesale destruction of the crop of every farmer in Britain? Famine? Of course not. We're lucky enough now to live in a part of the world where such an outcome is unthinkable. And why would the destruction of our food supply not lead to a food shortage? Efficient trade.
If you think trading produces no wealth, ask yourself how these shiny new buildings which are constantly going up in the City are being paid for. It is simply not possible to carry out projects like that on the back of moving some money back and forth: you have to create wealth to spend that much of it.
Sure, crashes happen, but the thing to bear in mind is that the population became generally wealthier from 2000 to 2013, regardless of 2008 and 2010. Indeed, we've had scores of terrible crashes over the last hundred years, yet we still live in a far wealthier world than existed a hundred years ago. The crashes are blips, but the overall net progress of the system is still good.
Monday 24th June 2013 21:28 GMT Marshalltown
Ah the joys of theory vs reality
It is true that in general the world's population is on average better off now than 100 years ago, yet the divergence between mean and median has increased and the median has moved leftward steadily releative to mean income. Curiously though, the same basal level of poverty is still the start point for that spread. That is, the poorest people to day are precisely as poor as they were a century ago and precisely as well equipped as ever. You need only look at the poor in Indian cities, the "troll" populations lurking under bridges and overpasses in the US, the slums of Rio de Janeiro to verify that the truly poor are in fact not better off now than they have ever been. Dirt poor remains dirt poor and has seen no "relative improvement." What really has changed is the spread between the wealthy and poorest. In the US the "one percent" is largely a product of those "efficient markets." Likewise, the wealth "created" really is not primarily goods. The fractional reserve banking system permits banks to issue money. Not the government - banks. If you worry about inflation look to your bank and the interest it charges vs. the interest it pays you. That spread really is inflationary.
Tuesday 25th June 2013 08:53 GMT Squander Two
Re: Ah the joys of theory vs reality
Leaving aside the example of people living under bridges in the US (since everyone I've ever known who works with the homeless insists that the root cause is usually mental health, not money), you're talking about people living in slums in undeveloped countries. But here's what I said:
> Look at countries that aren't properly involved in exchange trading. Compare the difficulties faced by their populations with ours.
You seem to think you're contradicting my point, but, as far as I can see, you're making it.
I do have to disagree with your explication of the statistics, though:
> the divergence between mean and median has increased and the median has moved leftward steadily releative to mean income. Curiously though, the same basal level of poverty is still the start point for that spread. That is, the poorest people to day are precisely as poor as they were a century ago and precisely as well equipped as ever.
This is a very long-winded and pessimistic way of spinning "Yes, most people have got wealthier, and the number of people in extreme poverty has decreased."
And "precisely as well equipped as ever"? So, no antibiotics, then? Just as likely to die of cholera or bubonic plague as ever?
Tuesday 25th June 2013 10:19 GMT Gerardo McFitzpatrick-O'Toole
Re: Fascinating article
Don't fail to realise that the reason there is so much money and brainpower involved in financial markets is that they serve to increase the productivity of practically every human endeavour by helping supply meet demand. If a year's work of a nuclear physicist results in a 1% increase in the world's cereal harvest in order to meet an anticipated future need, then that is time well-spent.
Algorithms and systems only make money when they are correct, and implemented efficiently compared to the competition. High-Frequency Trading is just the next logical step, with the potential to pick up information and transfer it to the market much faster, and to make or lose money at a correspondingly increased rate. Remember, traders only make money if someone else, somewhere, is also having their life made better in some way.
Friday 21st June 2013 12:02 GMT Pen-y-gors
Radical solutions needed
We need to go back to the root of why we have stock markets. They exist so that companies can raise money from other people for expansion, investment etc. The investor is (hopefully) paid a worthwhile return (dividend) on the investment based on the increased profit of the company. If things go well then, after a period of time, they may be able to sell their share in the company to someone else for more than they paid for it - although of course this doesn't actually benefit the company in any way. Arguably a proportion of any profit in trading shares should go back to the company.
But the whole point of a share market is to allow companies to flourish, not to provide an alternative to roulette, the gee-gees, mugging or visiting sub-post-offices with a shotgun. Share investment should be a long-term business, so there really needs to be a return to a slower system so that these cowboy gamblers are removed from the market. A simple solution is a) a small tax (0.5%?) on every transaction - irrelevant over five or ten years, and b) require that all trading is physical - shareholders must have physical cerificates, in their name, issued by the company, and must be in physical possession of the certificates at the time they sell the shares.
Not that difficult, is it?
Friday 21st June 2013 12:16 GMT DragonLord
Re: Radical solutions needed
Erm, I was under the impression that the original point of shares was to spread the risk of trading vessels so that the loss of 1 ship wouldn't bankrupt anyone. However equally everyone got a share of the profits when the ship arrived safely. The stock markets were borne out of the desire to get out of a venture if something came up.
Friday 21st June 2013 14:13 GMT SysKoll
Re: Radical solutions needed
Ooooh, look, a problem. I know! I hjave a universal solution (roll drums): a *new tax* ! That's right, ladies and germs, the tax, this good ole' cure-all, worshipped by English departments and arts majors everywhere, can cure your ills faster than you can spell n-i-n-c-o-m-p-o-o-p.
Taxes aren't a solution, they're a way to compound a problem.
Friday 21st June 2013 14:37 GMT Anonymous Coward
Friday 21st June 2013 18:27 GMT Anonymous Coward
@SysLoll - Re: Radical solutions needed
No taxes -> no revenue -> no services. I can't put it simpler than that, even if it is for you.
Are you trying to tell us we should abolish all taxes and start providing ourselves our own police, justice,fire, medical etc. services ? Care to explain how would you see police, military and justice turning their activity into a for-profit one ?
Friday 21st June 2013 19:24 GMT Intractable Potsherd
Re: @SysLoll - Radical solutions needed @AC 18:27
Please don't feed the libertarian. He'll come back and tell you that law and order isn't actually needed, it can all be dealt with as civil law, with paid-for services. He'll also say that, again, you can pay for the best contract for you for fire (remember the first fire-services (in the UK) were provided by insurance companies, and if you didn't have the right badge on your house, the firemen would let it burn to the ground). The military, equally (the libertarian will say) was run on the basis of mercenaries for far longer than the standing-army system, and that it prevented wars (though a look at the history of Europe between, say, 800 and 1800 proves what bollocks that is).
in the anti-tax libertarian (actually, anarchist) point of view, there is no such thing as a free lunch, and taxation is all a way to keep the weak poor living off the deserving rich , and ... oh, I can't be bothered any more. For an idea of how such a system might work (if you have money), have a look at Heinlein's "The Moon is a Harsh Mistress"
Monday 24th June 2013 10:28 GMT David Hicks
Re: @SysLoll - Radical solutions needed @AC 18:27
>> have a look at Heinlein's "The Moon is a Harsh Mistress"
Ah, "The Moon is a Harsh Mistress". Not so much a novel as a political rant with a few one-dimensional characters chucked in.
Did nothing to convince me of anything other than that Heinlein is overrated.
Monday 24th June 2013 11:01 GMT Squander Two
What are we trying to achieve with this hypothetical tax?
There are two different arguments being conflated here. The reason some people are proposing a new tax in this context is not in order to raise more money for the state but in order to disincentivize banks and investors from behaving in certain ways. This needn't necessarily lead to a greater tax take, as a transaction tax on banks could be offset by tax breaks elsewhere. So bringing up skoolznospitals is a straw man. After all, if the argument in favour of a tax is that we approve of the way the Government spends the money, then that same argument can be used to support any tax -- such as, for instance, a 4000% tax on contraception, or an increased license fee for pubs who allow black people in.
Taxation is about balance: we accept the effects of a certain amount of money being taken out of the economy in return for getting some stuff that we want the Government to do. But we're weighing up pros and cons here, not simply asserting that there are no cons. After all, if taxation doesn't cause any problems for the taxed, why not set it 100% and be done?
Friday 21st June 2013 16:02 GMT IHateWearingATie
Re: Radical solutions needed
While I agree with your point about financial services going serious off piste from it's proper role in a capitalist society, a financial transaction tax (FTT) is not the way to bring things back to sanity. Unless you manage to get full global co-operation, then the trades just switch jurisdictions (see Sweden's attempt at this previously) - proper global support for a FTT is harder to get than global support for other changes that will do the job just as well.
Oh and for the HFT supporters out there screaming 'market liquidity, market liquidity' you're wrong - HFT only increases market liquidity in the times where liquidity is not a problem, it doesn't magically increase liquidity when it is drying up because the same fundamentals apply to HFT as do the 'normal' market.
Friday 21st June 2013 16:47 GMT Tom 13
Re: Radical solutions needed
Small tax on what exactly? The total value of the transaction? Or just the profit from it? What if the transaction results in a loss? Can a loss be used to offset a gain? Over what time period?
Under the current US tax regime, I think the answer is a bit simpler. Go back to using the time a stock is held to determine it's eligibility for the capital gains tax rate. If it hasn't been held for at least 6 months, it gets taxed as income rather than capital gains.
I also don't see the need to have physical paper for the certificates, and especially don't see the need for the person who owns them to keep them. Many years ago I dabbled in the market, never had a certificate. Lost my shirt of course and haven't been able to get back into at the personal level since. But I held all the stocks I bought for more than 30 days. So that doesn't actually get you where we need to be.
In fact, I'd say right now the biggest problem with the stock market is something they keep selling us as a Good Thing(TM): Mutual funds. Granted the intended idea was a good one. But the way it has actually worked out is horrendous. The volatility of the market isn't just down to HFT, it's also down to too few people trading too many dollars. Even the richest person on the planet doesn't throw around the kind of money a mutual fund trader does.
Monday 24th June 2013 11:31 GMT Squander Two
"We need to go back to the root of why we have stock markets."
You are confusing why we have stock markets with why we have shares. They do not serve the same purpose (though there's some overlap). Stock markets mainly serve the secondary market in shares. Companies raise money and spread risk via the primary market.
Friday 21st June 2013 14:08 GMT Anonymous Coward
been making designs for these people,
they move servers so they are a few nano seconds close to the exchange feeds,
employ special hardware to search and trade,
each company has their own algorithms to try to spot and beet the rest,
I'm amazed the exchanges don't jump around more with all that goes on,
sooner or later , all the dominoes are going to line up and they total system will spiral down,
every one knows it, but wants to make money while they can,
Friday 21st June 2013 14:10 GMT Anonymous Coward
What Cantor’s e-Speed taught me about the frailties of Algo Trading (HFT)!
Speaking from experience here's three more reasons why Algo Trading / HFT's mess up. I was working at a large bank for US interest rate traders once. As an example, the traders wanted a mechanism to move large blocks of 30 year bonds without moving the market. So the goal was to design a black box (BB) to move small parcels throughout the day, trading on defined limits of the market at that time. The latter phrase proving crucial here!
The problem is: whenever its a busy morning the BB is reliant on timely receipt of MESSAGES and the timely order of those messages from the exchange’s trading systems, which is itself a kind of BB. And all of this withstanding the usual glitches like network latency, bottlenecks, outages etc What I was discovered was :-
#1. You can't recreate a live market in Beta. Why not? A. Because the beta system is not anything near as liquid, B. the beta is often not even running the same Exchange software version, and C. your black box is playing with other robots-- and not humans and robots and real-world actualities.
#2. There are significant subtleties in the way a busy morning can affect the ORDER in which your black box receives messages into its queue! I found cantor's e-Speed system sent messages out of order making it tricky to pair up last trade prices with current market pricing. It meant trade confirms with the actual SIZE moved were delayed well beyond where the market was now. In short, my BB frequently found itself in ill-defined state.. What to do next.…? Where was the market now? How much size was actually executed? What should the next BID / ASK posting be?
#3. The exchange provider only shares some technical subtleties with its clients, that is-- unless your are one of its darlings that generate the most fees: hedge funds and Goldman, Deutsche et al... What this means is that you are often operating in the dark. In addition the Exchanges systems can be the actual cause of the glitch. Without warning Cantor re-numerated trader operator ID’s overnight. Anyone who hadn’t re-logged on was unwittingly now using the wrong trader ID. Boy, that was messy! Traders getting another traders confirms, trades executing in the wrong books etc.
Friday 21st June 2013 14:10 GMT Anonymous Coward
HFT & Feeble Internet Security....
We're seeing more and more direct effects of hacking. How long till a hedge-fund with a HFT system gets an insider to 'hack' a highly watched Twitter account? Oh wait, didn't that happen! "Hacked AP tweet claiming White House explosion causes Dow dip" :-
Friday 21st June 2013 14:40 GMT bigtimehustler
Saturday 22nd June 2013 08:29 GMT Anonymous Coward
Not that I can't really be bothered now (it's Saturday afternoon in my part of the world, I have beer in the fridge, meh), but electrons move somewhere like millimeters per hour, and in the opposite direction of what we call current, which, as already mentioned above somewhere, travels at between 65% to 80% or so of the speed of light, depending on what kind of medium (UTP, STP, Coax, etc.).
And if that's if you are talking DC. If the "signal" is AC, effectively the electrons just jitter around the same place not going anywhere fast, like my friends last night in the local pub.
Back to beer
Friday 21st June 2013 14:45 GMT Anonymous Coward
Regulators Beware: Inadequate testing or Impossible Simulation...
Having worked on HFT systems, you would be amazed at how many changes are made on the fly without care and consideration to investors. Think FIAT: Fix It Again Tomorrow as a working mantra. Part of the problem is this. You can't recreate a live complex organic trading system in beta. Often the beta isn't even running the same release as the live system. So it isn't just a question of numbers i.e. only testing 10% of the orders. Its also a question of real-world complexity in the way orders are placed and in the complex interaction between all the different players.
Complexity is exponentially increased with automated market makers (AMM) and their interconnected exchanges, HFT systems, hardwired bank and institutional screens, retail systems, and legacy phone or pit orders etc. When you sprinkle in AMM stock-pinging, liquidity-rebate-trading, front-running, fat-finger trades, legitimate cancelled orders, and competing orders from co-located servers versus those at a distance... life can get very messy.
Its very difficult to build a good simulation. I wish the regulators would acknowledge this and herald it as warning.... Instead they continue to see these events as one-off problems in tunnel vision fashion. So when-is the next flash-crash or IPO non-event...? ...FIAT!
Friday 21st June 2013 14:46 GMT Anonymous Coward
Not looking to fill a position in the Spelling Police but please see square brackets ....
The design choices made in chips today and architecture of how these systems are laid our [OUT] aren't optimised for this type of behaviour,"\\
will spend a ton on the expertise required to write its software (and the pricy coloation [co-location] fees with the exchanges).
However, just like cloud computing, they may be [be] more similar than dissimilar
HFT job adverts typically ask for C++ and/or Java skills, along with a familiary [familiarity] with Unix systems.
execute buy and sell orders in as low [little] a time as possible
Friday 21st June 2013 14:58 GMT Dave P 1
Time to revert back?
Perhaps it is time to revert back to a paper system. Print out a buy/sell order and submit it to be posted. Pay for your purchase; no margins, it will be cash on the line. You either own the asset outright, or you do not; keep it simple and no short-selling. Buy for the long term; not milliseconds. Sure this will ruin the industry. But it is an industry that adds no value.
Friday 21st June 2013 15:46 GMT Stevie
According to an NPR report on this some two weeks ago (shame on El Reg that let a not-for-profit analogue radio network whose major focus is politics, humour and classical music beat them to press: call yourself an IT "news" site? Don the Cone of Shame!) the problem is orders of magnitude worse than reported here. Trade battles happen in the space of microseconds, not seconds, and no-one is thinking of the children.
Who will get stuck with the bill for this idiocy? The very people who are supposed to benefit from the stock market - human investors. Mostly (as usual) those with their pension plans woven tightly into Hedge Funds, those sinkholes where one can "safely" stash the results of digital stupidity and malfeasance. See: Naked short selling (which isn't as exciting as you might think but the hip name for something that in any other milieu would be termed "criminal fraud").
The real issue here is that those writing the software do not have the greater good of the market front and center in their problem horizon (and probably have little interest in the business of stock market economies per se) so eventually they *will* drive the whole thing onto the rocks (again) while convinced that they are doing a good job.
Friday 21st June 2013 16:03 GMT Stevie
I'm puzzled. Since the word "algorithm" is used throughout the article, why the occasional salting of "algo"?
Quick recheck: "algorithm" is used in quoting a source, "algo" when the author is using his own voice.
Hmm. So, is the author simply trying to sound more like "one of the lads"? If so, I should point out that I've never heard the word "algo" used by any of the seriously clever people we keep around here to write them.
So, I would like to ask the other commentators: Is "algo" a clixby term or a clabby one?
Monday 24th June 2013 10:14 GMT Anonymous Coward
As someone in the industy, algo is the usual term that we use when discussing this stuff. Why use 4 syllables when 2 will do? You only use "algorithm" when spelling it out to someone who doesn't know what an algo is. Which is presuamably why the quotes all use it. The author's use of "algo" would hopefully mean he isn't talking down to a presumably technically literate audience, who can be assumed to be capable figuring it out from context if necessary, or at worst googling it. Or he's trying to come across as someone who is in the know to make himself sound more knowledgable, if you're feeling cynical.
Friday 21st June 2013 17:21 GMT Petrea Mitchell
Market signals: Do they really exist anymore?
Here's what I've been wondering ever since learning about how common HFT has become: Company stock prices, individually and in aggregate, are seen as a very important piece of economic information that carries a signal about how well a company or an economy is doing. Given how much market activity these days consists of algorithms chasing each other around, is there any point in still seeing stock prices as an indicator of anything regarding the economy outside the stock exchange? If not, what indicator(s) do we replace them with?
Friday 21st June 2013 19:49 GMT Anonymous Coward
I have not read more rubbish being posted by people on this. Normally people who post comments on The Register are erudite and knowledgeable. Not in this case, and hence have created an account to post.
Firstly - the speed of light is not related to electrons going down copper wire. They are related to the propagation of the Poynting Vector. This is at the speed of light. Someone posted that the drift velocity was slow. That is true. Well said.
Someone else said that Banks do not generate wealth. Please can you at least study some economics before you trot out your Daily Mail comments and pollute the lovely register with incorrect facts.
Banking creates wealth in two ways. Firstly it enables lending of money so that time value can be realised. Time value of money means that I am willing to pay more for money today, than in the future. This gives rise to interest rates and means that people can lend their excess money to other people. This has enabled the building of most things you can think of. Banks can do this due to fractional reserve banking, and means that as long as a bank can cover its creditors, it can lend more money out. This creates more money in the economy and means that you as an Apple fanbois can go and buy an iPad to read the Daily Mail.
The second thing that banks allow you to do is transfer risk. This generates wealth as it means risk can be transferred to people who are willing to take the larger risk and hence de-risks a company that going to build a hospital, a new iSomething or a tank and transfers it to speculators who are willing to take the chance. Some win some loose. Welcome to real life.
Stock markets are a example of the first, derivative markets are an example of the second.
HFT is an interesting area. When people traded by carrier pigeon and did a trade a week, a telegraph meant that the person was now HFT at 100x a day. To say HFT outright is wrong is just misinformed. It is however not as simple as that, as the issue does arise when two things can happen. The "flash crash" meant that the exchanges cancelled a load of orders made in good faith. That means that some people would have lost the money that they made (HFT or not). How can a single body choose whether a legal agreement should be upheld or not? Sounds wrong to me.
The second reason that HFT raises eyebrows is around the concept of liquidity. Liquidity is how easy it is to buy and sell something. The problem is that the market looks very liquid when HFT algos are running, however as soon as they are switched off the liquidity dries up. This can have unexpected effects in terms of market movements. You think you are holding something that is easy to sell, and suddenly the algo is switched off and you are holding something that on paper is worth a lot, but no-one will buy it.
Transaction tax is not a good idea and will cause all sorts of instability in the market. Basically it would drive the trading volume to places where there was no transaction tax transferring wealth to those jurisdictions. It also does not solve the problem of liquidity that I discussed above.
I am not saying support HFT, but I am saying that if you are going to critcise it, at least have a basis for the criticism.
Posted as an Anonymous Coward due to where I work
Friday 21st June 2013 21:00 GMT Herby
High Frequency Trading vs. SPAM?
It seems to me that these appear similar. They BOTH rely on the low/no cost of transactions. The SPAMmer basically has a zero cost method of emitting junk messages (at least until he is caught), and the HFT guy has little cost in his transactions other than the cost of the security itself (which goes up/down by mere pennies). Sure there is a hardware cost, and for both people there is an ever escalating "arms race" that they need to compete in, but the actions are quite similar.
Now if there were a cost per transaction (however small), that would make the HFT guys think before trying to eek out the gain of a small amount (read cents) on a stick price the frequency might lower as the margins need to be better to recover the cost of the trade. This could be accommodated with a minute (.01% maybe) price. Another option might be to only allow "publishing" of trades on one-second intervals. Plenty of time for average people (like me) to see what is happening, but for the HFTs it would be an eternity, allowing everyone to catch up with the information at hand.
As for spammers, all we can do is hope that their economic model will eventually NOT make any sense. For the moment, it does, as maybe .001% respond but that is enough when you send out blasts in the millions of messages about enhancing drugs and the like.
Live & Learn
Monday 24th June 2013 11:14 GMT Squander Two
> the HFT guy has little cost in his transactions other than the cost of the security itself (which goes up/down by mere pennies).
Would you care to explain how a load of transactions which had virtually no cost, and therefore, according to your argument, no risk, caused a crash? Are you aware that a crash involves, by definition, the loss of a large amount of money? Some people might posit that the loss of money constitutes a cost and the potential loss of money constitutes a risk.
Monday 24th June 2013 18:01 GMT Anonymous Coward
"the loss of money constitutes a cost and the potential loss of money constitutes a risk."
Where is the risk when the global financial institutions have got themselves onto the "Too Big To Fail, Taxpayers Must Bail" list ?
Oh, that's right, the taxpayers and other less fortunate members of the public get the risk (and the austerity for years to come, and the demolition of the former welfare state, health service, education system, etc), and the City gets the benefits.
Nice work if you can get it.
UK plc. K is for Kleptocracy.
Tuesday 25th June 2013 09:10 GMT Squander Two
> Where is the risk when the global financial institutions have got themselves onto the "Too Big To Fail, Taxpayers Must Bail" list ?
Well, that's a different matter. For the record, I absolutely opposed the bail-outs then, and I still do, for precisely that reason. And I saw a bit of interesting bail-out stuff from the inside which, I don't mind telling you, was absolutely *REDACTED BY NSA*.
As far as I could see at the time, the bail-outs were largely government-led. Once the governments had announced that they were going to bail out banks, the banks would of course have been stupid not to clamour for the money. An appalling kick in the teeth for every small business that had ever failed. "If you screw up, you lose. If they screw you up, we give them your taxes."
And don't get me started on a "Labour" government announcing that they're going to tax working people, give the tax to banks, and then tell the banks to lend the money back to the same people it came from, but at interest.
Saw in the news the other day Stephen Hester referred to as "the saviour of RBS". Er, no, I thought, that would be the taxpayer.
Friday 21st June 2013 21:02 GMT John Smith 19
So how many of those guys in stripy shirts are *left* actually dealing stuff?
With 80% in the US and 40% in the UK it sounds like a hell of a lot of the high volume stuff is being done by the machines.
And (it appears) they can be fooled easily
I'd suggest what really stuffs the market is the high frequency trading and high frequency cancellation of those trades.
Not to mention that positive feedback phenomena which can make trading so exciting.
Saturday 22nd June 2013 00:14 GMT Zola
Wow - an article on black-box trading and no mention of...
Prediction Company, whose system is based on chaos theory and used by UBS since the early 90s, and still in use to this day?
The book "The Predictors" (by Thomas Bass) details how it all came about, and is itself a sequel to his earlier book "The Eudaemonic Pie" (British title: "The Newtonian Casino") which tells the (true) tale of the same protagonists (then as university students in the late 1970s) using chaos theory and wearable computers to beat the Las Vegas casinos at roulette.
The first book is by far the more entertaining of the two.
Saturday 22nd June 2013 12:37 GMT Anonymous Coward
Saturday 22nd June 2013 18:31 GMT Mark Gannon
Read 23 things they never told you about capitalism...
A great book called "23 things they never told you about capitalism" Ha-Joon Chang should resonate with a lot of the posters here. One of his very well made points is that markets need to slow down; their purpose is to provide liquidity to firms for long term investment...hyper speed trading is about as far from the world of real world companies as you can get. Given the 2008 crash surely everyone should be horrified at the continuance of these trading practices.
Monday 24th June 2013 08:32 GMT Tom 7
Re: Read 23 things they never told you about capitalism...
The crashes will still occur even without high speed trading - they are simply the result of the positive feedback in the system where a small percentage of the shares sold dictates the price of the rest so a small trade can produce a large amount of virtual money. This leads to stock having value much greater than was ever paid for it. This is used as collateral to get more money which is used to pay themselves huge bonuses and then when there is a shock to the system then the whole edifice collapses but all the money used to buy stock has been shuffled abroad out of reach so when some dickhead thinks the whole thing is worth rescuing we have to screw the only people with real money - the taxpayer.
What is great about high speed trading is it gives an illusion of economic growth - I sell you £1m in shares the economics says that contributes £1m in trade. You sell those back to me and me back to you a million of times and that's a £2billion added to the economy. So governments encourage this so their view of the economy grows and while money is of dancing back and forth fuck all is invested in making things - so we are all richer and richer but we just have less real stuff. If it wasn't for the fact that I can buy 1,000,000,000 more transistors for the same price I could twenty years ago I'd be convinced the economy had really shrunk in that time.
Monday 24th June 2013 11:22 GMT Squander Two
Save us from the central planners.
> their purpose is to provide liquidity to firms for long term investment
No, Chang doesn't get to define the purpose of markets, no matter how much he wants to. Markets are simply the fact that people trade with each other. People trade for lots of different reasons. There is no single overarching "purpose" of markets. It's just people interaacting. Markets have some good effects, and some of us humans have purposes when we use markets, and politicians might have a purpose in mind when they do something to enable or restrict markets, but the markets themselves have no purpose. They evolved.
> Given the 2008 crash surely everyone should be horrified at the continuance of these trading practices.
Why, when no-one has ever claimed that the 2008 crash was in any way caused by HFT?
Sunday 23rd June 2013 16:14 GMT dajames
Tuesday 25th June 2013 08:57 GMT Squander Two
Re: Wouldn't it be nice if ...
> Wouldn't it be nice if stock market transactions were subject to a 14-day cooling off period
You mean I can buy a controlling interest in a company, sell off all its assets, lay off all the staff, and then force the seller to buy back the shares at the same price I bought them for a week earlier? Fantastic! I'm in.
Monday 24th June 2013 09:31 GMT Zippy's Sausage Factory
Monday 24th June 2013 13:52 GMT Mark A
It seems like a problem that speed (and FPGAs) means the algorithms have to be simplified, potentially removing the possibility of any reasonable safeguards being built in. If everything was throttled to a slower speed, that headroom could be used for smarter rather than faster algorithms with better stability.
What puzzles me is why don't regulators enforce this? Seems like it could achieved with a bit of buffering into the exchange with a random shuffle (based off an entropy source). Or is everyone making too much money leasing the fastest slots/lines?