Slashdot Mirror
HFT Nothing To Worry About (at Least In Australia)
angry tapir writes "Although software-driven high-frequency trading has got a pretty bad rap (being blamed for the so-called 'Flash Crash' in 2012 for example) Australia's chief financial regulator ASIC says that, in Australia at least, it's not cause for concern. After an in-depth study of HFT in Australian markets, ASIC decided to hold off on previously considered regulatory changes (such as implementing a 'pause' for some small trades)."
Why would ASIC be concerned about software-based traders? They know that, while it renders them somewhat inflexible, they are both far faster and substantially more power efficient by doing it in hardware...
I'd like to see HFT banned, or taxed, or slowed down in some way, just because the big traders use it and their millisecond advantage over the non-insiders to steal a small percentage on each trade. They amass billions by siphoning it away from the majority of people in the market, and in return give us nothing of social value.
*Two* HFT stories on Slashdot's home page? Well played editors, well played -
This a pet topic for these guys?
http://news.slashdot.org/story/13/06/18/0257224/have-we-hit-peak-hft
HFT isn't a system stability problem as much as it is an access problem. What it does is increase the cost of entry into the market -- those who don't engage in HFT wind up paying for those who do, and so it winds up penalizing people with smaller portfolios and shifting the costs of it onto them. What you need to understand about profit is that it is always at the expense of someone else. And HFT is the sublime example of how to nickle and dime the less fortunate to death. These fractions of a penny here and there add up because it gets compounded by interest rate. Over time, the spread between those who have it and those who don't will grow; As is the trend in any investment-based system.
#fuckbeta #iamslashdot #dicemustdie
I think the opposite would actually work better.
If the official rules stated "HFT is totally *un*regulated --- feel free to run your buggies, most insane, glitchy, and flawed HFT software" --- immediately all the other HFT software systems would be coded to watch for crazy non-justified buying&selling.
With all this regulation, if one bank's trading software starts going insane, the other banks start following them (and creat a positive feedback loop) under the assumption that in such a regulated industry the insane software must know something. If it were further de-regulated, the other software would assume the other software was poorly coded, and basically LOL at the bugs and profit from it until someone pulled the plug on the bad algorithm. And with that risk - I imagine a *lot* of interested would be in automating such plug-pulling checks so they happen in a very small number of miliseconds so the market can't crash too far before the kill switch hits.
If there were a transaction tax on stock market trades, that would eliminate whatever advantage there is. These guys make money on low margin high volume trading. Just about any transaction tax will make those low margins disappear.
The disadvantage is trades might now take minutes instead of seconds due to decreased volume. But maybe that's not a bad thing.
.. in the brains of an engineer:
https://en.wikipedia.org/wiki/Application-specific_integrated_circuit/
Can people remember beyond 1 year? Is it cultural that some areas have less memory than others?
Rather than address the obvious fault in the parent's post, I'll suggest another problem:
War. Not necessarily all out war, but hacks or even directing large corporations under the control of the enemy could screw up such a system much easier than the old fashioned methods for undermining an enemy's economy.
Long term, it would be wise to join the Chamber of Commerce and help them in promoting the destruction of the USA - it is slow but it'll do the trick.
Democracy Now! - uncensored, anti-establishment news
because when I read "ASIC decided to hold off on previously considered regulatory changes", the first thing I thought was somebody got bribed.
Austrailia has a world-wide reputation of being laid-back, easy going, and - sometimes - incredibly rational (real gun laws in response to mass killings).
But on the other hand, they keep electing right-wing governments more than willing to be trained poodles for US corporate and foreign policy.
If you have billions in capital, it is extremely hard to move around billions in assets without all the small traders taking notice, and piling on before you can reach your full position. That's why large traders like Buffet absolutely hate day traders, and has never split his stock, causing shares in his company to be valued at over $65000 per share last I checked. Being able to trade freely and quickly is one of the few great equalizers in large capital markets.
So as an individual with say.. $2-$5k to play around with.. how do I get in on this HFT thing?
It's the latency afforded by buying rackspace directly within stock exchanges that poses a problem.
Lets just assume 100 shares and say player1 is asking $10.04 and player2 is bidding $10.00, and there are no other players. For starters, I don't really understand how a transaction happens at all here. It would seem that one of these two would need to modify their bid or ask. So now lets say you are a new player3 and are employing high frequency trading. The most obvious thing that comes to my mind is that you could get in the middle of this and buy at $10.04 from player1 and then sell at $10.00 to player2 and lose $0.04 cents. Clearly this is no good. Alternatively player3 could place a bid at $10.02 and hope that player1 will lower his ask. If this transactions occurs then player3 can start asking at $10.03 and hope that player2 will now up his bid. If this goes as planned player3 makes $0.01. There would be a window of risk here for player3. The stock has been bought at $10.02 and the only buyer is bidding $10.00 and the ask has just come down from $10.04 to $10.03. There is no guarantee there will ever be a buyer at greater than $10.02. If this what actually goes on, then player3 is sort of facilitating a transaction. The spread was originally too wide for either player1 or player2 to move. player3 has essentially lowered the spread and enticed both player1 and player2 to make a transaction. The above is really sort of a big question. I have no idea what HFT is really doing.
No, it's even stupider than that.
It's not anti-capitalism, it's anti-markets. Even if the proletariat siezes the means of production, this guy's statement will be sitting there, making the proletariat to be a bunch of bad guys whenever their communist factories sell anything they produce.
It's so stupid, that your brain failed to believe how amazingly stupid it was. It's like you're standing next to a wall, and you glance at it, and see a ten-foot-around circle. "Huh, a big circle. Weird. Whatever." But I look at the wall, and realize: that's Cthulhu's pupil. You're not to blame. You're not stupid for thinking that's a circle on a wall. I don't fault you a bit, and I prefer your world. But you're wrong. Just like you were wrong about the magnitude of how stupid the GP is.
Move from the current "income" and "gain" tax basis to a gross receipts tax. Every transaction is a taxable event - retail, wholesale, personal, business. It turns out that the rate would be exceptionally low - 2-3% tops. It would result in larger markups on long supply chains, make day trading very costly from a tax perspective, increase the cost of most items at the retail level, and burden corporations - especially holding companies and multi-layer shell corporations meant to shield/dodge local taxes and reduce liability. OTOH, it would reward short supply chains (local suppliers would have an advantage over distributed goods), long term investments, and punish multi-layer shell corporations meant to shield/dodge local taxes and reduce liability.
I find it interesting that investment brokers, real estate agents, copier companies, energy supply companies - nearly everyone in business - basis their price on the cost of service or a fraction of the gross transaction, but the government only charges you based on your profit. Can you imagine if you offered to pay electric bill only if you made a profit, and you wrote the rules to minimize what constituted profit?
Is it just my observation, or are there way too many stupid people in the world?
But with that neophyte error you've shown that you're fucking clueless.
Unfortunately following our Great and Powerful friends into whatever folly they get themselves into has been Australia's foreign policy since federation.
Plan My Week for iPhone
1) New previously-unexploited market opportunity is discovered/developed (e.g. HFT).
2) Initial enterers into said market start making bucketloads of money.
3) People complain and say this should be regulated, its not fair to X, Y, and Z, this is a plague on society, etc.
a) This of course conveniently ignores the fact that when trades are done voluntarily they are only done because both parties think they will benefit[*].
4) If we're lucky and the market doesn't get regulated too much, more people enter said market, reducing the profit margin while providing more benefit to people in said market. Eventually there isn't much money to be made in it at all and it's just another product/service everyone takes for granted.
b) If we're unlucky then the market gets regulated too much, stunting its eventual development of another taken-for-granted product/service, leaving it prone to disruption whenever the regulations change.
[*] People above have already given many reasons why HFT ends up benefiting the market, for example punker's post.
ASIC declining to regulate HFT shouldn't be considered an endorsement of the practice. As it is, they are flat out trying to root out insider trading and abuses of Corporations Law. (e.g. Trading while insolvent; embezzlement etc etc). It doesn't have a great reputation on either front. There's the odd successful case now and then, but that seems to be the exception rather than the rule.
Having said that, I'd rather have it than not, but it would be nice if it could be made a little more effective.
Plan My Week for iPhone
Thanks for the informed reply. I'm aware that the crash you mentioned and the crash I mentioned are not one and the same. The link you gave provided little insight into its cause. Do you have any more information about it? The crash I mentioned wasn't due to a fat-finger trade either, just a really large trade placed as a hedge, but done in a time of low liquidity.
As for what I would consider HFT: it is a bit of a grey area and hard to draw a precise line, for which I'm sure you'll forgive me, but as I see it, it is entirely independent of the "why" the trades are being made, just the timing, and is a bit of a relative thing: how fast a trade is turned around, how new the information must be to act on it, and so on, somewhat relative to what everybody else is or had been doing. Things like holding a stock for 10 seconds before selling it, getting your orders in before anybody else, and so on, and doing this regularly and as part of a business model--these things I'd call HFT.
My focus is on the underlying mechanisms the trader/algorithm uses, not any kind of official designation or requirements. Automated market making would therefore in my opinion fall under this category, for most liquid assets. (For less liquid assets, trades are likely slower, so it'd be something I wouldn't consider high-frequency.) Officially designated market makers, under this perspective, are simply HFTers employed with a specific algorithm (make trades at the quoted prices) and specific requirements (always stay in the market / provide a minimum amount of liquidity at all times). In exchange for these requirements, they may get certain benefits such as liquidity rebates or better information.
I would like to point out that much of the debate as to whether HFT increases or decreases volatility hinges upon what exactly you consider to be HFT. If you take market makers and arbitratrageurs out of your definition, if you even can (you pointed out the difficulty here), I believe most of what is left is going to be those increasing volatility. And then you could say HFT increases volatility, or removes liquidity. But doing that seems a bit arbitrary to me. If you look at it from this perspective, I think you can understand why people make the claim that HFT reduces volatility or adds liquidity. How would you define the line and why would you put it there?
Comment removed based on user account deletion
In 2006, BHP (one of the busiest tickers on the ASX) saw an average of around 1.6million trades per day (20 non extraordinary business days).
Yesterday the average was ~72k trades. Furthermore the ADV has dropped by 80% over the same period.
HFT only works if there's serious momentum in a ticker's price direction (up or down). The HFT systems detect the momentum before anyone else and take advantage of it - if no one's trading there's no momentum to be had hence no profit - worse still is that if the HFT corp is a market maker or similar, they may have to take a loss so as to meet exchange obligations so as to be entitled to rebates based on resting orders they placed which were executed.
The other way HFT works is by arbitrage. The ASX has a competitor called ChiX that lists all it's major tickers in a colo close to the ASX primary matcher (~3ms RTT colo2colo). The problem is that the spreads are near identical on both exchanges nearly all the time for all the majors, even though its cheaper to trade on Chi-X, so there's very few arbitrage opportunity for the HFT crowd to profit from.
It is pretty easy to do a google for "site:linkedin.com sydney" where is Tibra or Optiver or Liquid Capital or Boronia (Grinham) or SIG and notice how there's a lot more people listing those places as past instead of current (specially Liquid Capital)
A lot of these companies were making money hand over fist during the GFC (because of price volatilities) and nothing else (no special quant math based prediction models etc), now that the volatility is down, trade volumes are down and there's a lot more of them in the market, they're not able to make as much or any profits, hence the huge turnover in staff - Remember this is not an issue of mistakenly employing someone who can't do fizzbuzz (at 150K base +up to 30% bonus, they make sure they know what they're getting).
And that's why you see the likes of Tibra and co trying to hire like crazy anyone and everyone how ever did a machine learning or financial engineering course at uni - they think that if they can even slightly increase the sophistication of their trading strategies from the simple "quickly buy low then sell high" technique that it will allow them to get back on top, make profits and possibly help support Danny's dream of one day being the next 20-20 day/night Packer.
The link you gave provided little insight into its cause. Do you have any more information about it?
The way I see it it's about the structure of the trades. From their text above the plots:
The price dropped from $796 to $775 in about 3/4 of a second, then rebounded to $793 a second later. The drop invovled 307 trades and 57,255 shares from 10 exchanges + dark pools. During the drop, there were 5 orders placed for every trade executed (meaning 4 orders placed/canceled for every trade).
Having about 1.5k quotes posted in 0.75s (with 80% being canceled) surely shows that HFT algos were active and yet did not absorb 57k shares of GOOG (that's about 2.5% of the daily volume at the time). Which is not a thinly traded stock by any measure. It's not a conclusive proof of HFT causing a drop of 2.5% in less than a second though, but I don't think you can easily get such a proof without knowing who traded what. There is only so much one can glean from trade data at this level after all.
Due to some time constraints I'll beg to be excused if, in lieu of the rest of the comment I was going to post, I'll give you this example (originally from here but the harvard link seems unavailable) of a HFT technique that, shall we say, 'plays' liquidity to increase volatility. The problem, as I see it, is how to discourage people who have the means to use them in such a fashion. Bear in mind that HFT requires quite an investment in infrastructure and software, so voluntarily refraining from taking advantage of that infrastructure in ways that exploit its advantage at the expense of slower traders is not a believable option. However, I'll also freely state that I have little faith in simple-sounding solutions to complex problems, which is what most pro- and anti-HFT rhetoric brings. Fortunately, it is not my job to find the proper solutions :-)
HFT relies on the assumption that the real value of a stock changes several times per second (even down to the scale of the nano second). Everyone can see that the only event that can happen at such a high frquency is other high-frequency mechanism taking decisions. This is clearly a zero-sum game that runs on speculation.
There is a strong case for imposing a lower trading frquency. I would have said a one-hour timestep but some people go as far as proposing a one-day step. After all, stock exchanges close at night and during weekends.
Steve Jobs death also proved HFT to be ridiculous : Do you remember what happened to Apple's stock during this event? It was suspended for a day. Because everyone knew this was a signficiant event that would change the stock value but that with HFT people would be trying to anticipate short-term reaction (SELL! SELL!) rather than actual stock values. This is an admission that HFT actually hurts the ability of stock exchanges at discerning the true prices of stocks.
The Wise adapts himself to the world. The Fool adapts the world to himself. Therefore, all progress depends on the Fool.
The details don't matter when the entire point is to be a man in the middle attack.
Simply put, when two traders decide that they're willing to pay (perhaps) an extra 0.1% to ensure that their trade happens _now_, then that little margin is the _only_ reward offered to competitive (read: high-frequency) traders. If I don't want to be that huge sucker who pays the extra 0.1% for immediate service, then I will place a limit order. That limit order essentially boils down to using weak tools (broker's web interface, internet lag, etc.) to compete with HFTs and save my precious 0.1% to myself. The fact that HFTs exist means that my competitive limit bid or offer won't yield much over an immediate market order at any price. Grudges come from people with weak tools who pay too many big fees and consequently lose the day-trading game. Serious grudges come when whole exchanges cancel orders due to one party's HFTing failures (losses). If executed orders are cancelled too often, then some other exchange will try to complete with NYSE, AMEX, TSX, DAX, etc. by guaranteeing no cancelled trades.
http://www.lietaer.com/2010/03/the-worgl-experiment/ is implemented
Casteism