Occasionally but pretty rarely. By the time a trade has become actually risk free, as in "I can instantly buy X here and sell X there and make a profit", someone will have realized that and done the trade when it was almost risk free- the that the price is likely but not guaranteed to go up there- and made the trade first.
> Wall Street trading desks have benefited in the past four years from large market swings around the Covid-19 pandemic, Russia’s full-scale invasion of Ukraine and the decision by central banks to lift interest rates amid stubborn inflation.
So...by capitalizing on tragedy? (I know it's nothing new, it just makes me a little sad)
I mean negative moves tend to be more volatile than positive ones, but yeah I don't think "capitalizing on tragedy" is the right way to frame it. Financial conditions during bad news events would turn out worse if there weren't these market makers providing liquidity. Some pharma companies profited from COVID but it was because of valuable services provided. If they didn't "capitalize on the tragedy" we'd have a bigger problem lol.
Am I wrong in saying that high-speed trading does nothing to make the world a better place, and simply sucks up hardware, infrastructure, and talent to enrich a handful of people? If not, how about a per-transaction tax on securities trades, one that wouldn't be out of line for individual investors, but would extract considerable funds from HSTs that could be used for things that will make the world a better place?
Not entirely "nothing", as a result of HFT market making all participants are getting smaller spreads and increased liquidity. However... That has huge diminishing returns. The first few steps to electronic markets and fast market makers dropped the spread considerably (making trading cheaper for everyone), all the steps that came after cost a lot more and didn't really improve things much.
So they're not adding a lot now relative to the cost and complexity. But banning the whole practice will make the markets function much worse and more expensive for everyone.
The only agent the spread matters to is the HFT and a market maker so it’s a circular reasoning.
Similar for liquidity provision; most of the optimal control theory solutions for market making with alpha signal will widen the spread and remove liquidity the moment they sense something is off.
If I buy a stock and sell it three years later, my profit is reduced by the spread. If the spread is 5% that might be a significant portion of my profit.
So if a stock price crashes there's no one around to buy the stock, so HFT doesn't help in that scenario. They're not going to reduce the buy/ask spread.
But if a stock is going up, HFT inserts themselves between the buyer and the seller and takes a small cut in the process. I don't see how it helps the liquidity stock, but seems to be more like an HFT tax.
They didn't increase the number of stock shares, just got in between the buy/ask margin.
Going in between the buy/ask margin means the margin got smaller which is good for all traders in the market. There is no way to "go in between" and not make it smaller, because the best price gets the order.
Jane Street is not a huge player in HFT, they are primarily a liquidity provider and therefore actually do enable people to sell stocks during a crash. Any registered market maker has specific rules imposed by the exchange about always needing to provide a 2 sided offer within a reasonable bound of the current market.
It does seem like an incredible waste of resources of all kinds, but I'm not sure whether a tax would fix things.
This is a problem of market structures, if I understand it correctly; facilitated largely by Reg NMI in the US and its equivalents globally. And it's not like all was well before that: Instead of HFTs, it was market makers or specialists pocketing impressive, largely risk-free profits.
What you'd ideally want is to preserve the good that the HFTs facilitate (liquidity, more efficient price discovery, smaller spreads etc.) while avoiding the bad (siphoning off investors' money from the markets and math and physics PhDs from the world's top schools), but I haven't yet heard a convincing proposal for that.
One proposal I heard was to make the market "tick slower". Accumulate orders for, e.g., a minute and then resolve them all at once (without favoring the early ones). This way a delay of a few seconds doesn't matter.
Former options trader. This improves dealers’ profits.
Consider a once-a-day auction. The person who puts in an order at 9:35AM incorporates five minutes’ information. The person who does so at 3:55PM, all but five minutes’. Anything material that happened at Noon is extracted from the early order and benefited to the late one. If the orders are unsealed, the later order can incorporate the early order’s existence.
With options, that means the high-speed late order has better information about the underlying stock than the orders preceding it. So it generates risk-free profits from them.
What if the orders were sealed and the auction happened much more frequently, like every minute? It doesn't seem like late traders would get much of an advantage.
> What if the auction gets resolved at a random (uniformly distributed) point in time in some small time interval?
I'd put in lowball bids and ludicrous (but plausibly deniable) offers and wait for the unreasonably-short windows.
The solution to this problem is people not wanting immediacy. (Same with scalpers.) But people like immediacy. Unfortunately, there is also political capital in railing against those who provide that immediacy. In a way, it's a stable socioeconomic system: HFT provides liquidity, market partiipants take it, activists rail against it, regulators complain and win the activists' favour.
It was worse in the "good old days". The cost of trading was much higher and the people that felt that were the people who needed the market to make the world work (risk management, investment etc.)
Markets make the world a better place (it's not a new idea), and more efficient markets are better than less efficient ones.
> If not, how about a per-transaction tax on securities trades, one that wouldn't be out of line for individual investors,
Why do you want my pension fund to spend more of my money on trading costs?
Trading is a self-fulfilling prophecy the more trades of anything that can be conjured up out of thin air is seen as a ripe market to the financial industry.
HST turns what would have been a single trade into multiple trades making volume go up and to the right.
I've never understood why job creation would be considered something desirable in and of itself. Doesn't it make much more sense for a society as a whole to optimize for efficiency and then reallocate the now-redundant employees to tasks that actually need them?
And even disregarding that, practically, the kind of people working in HFT usually have a skill sets that make them highly employable anywhere. If anything, HFT is draining some of the best mathematicians and software engineers from other projects that can't afford to pay as well.
Who gets to define whats most efficient? And sadly, society grossly misallocates its resources when you think about whats right.
Its more important that we live in huge/new houses and drive new cars and have nice vacations than fund to finding a cure for cancer, and people make this decision every day.
If we have the option to make a human's job redundant through automation (without a disproportional reduction in quality), but choose not to because of "preserving jobs", that's inefficient.
In my view, it's disingenuous and cynical towards the person doing the job: They're effectively historical actors in a museum because we haven't figured out how else to solve for the distribution of monetary and "soft" values like societal status, feeling like a productive member of society etc.
There's nothing at all wrong in my view with letting people do what they want and cross-subsidizing that activity as a society, but why not call it what it is? And why stop at office-style work and exclude e.g. art?
You could say the same thing about most HN users' software jobs, but then upon deeper inspection, you would realize: oh, there's a value here (however slight), which is why people are getting paid for it.
Impacts of HFT on society:
- You no longer have to pay a stockbroker $100 over the phone to take a position and hope they get you a decent price. Trading is instant and spreads are extremely tight.
- Money they make is made by 1. offering you convenience (usually for an imperceptibly small fee built into the spread) or against people trying to do the exact same thing that they are (intraday trading).
The intraday trading is a zero sum game, but the liquidity providing is net positive. The zero sum component only affects you if you're trying to profit from intra-day trading too, in which case, you're a hypocrite for calling them out.
Amongst people who understand HFT well enough to realize it benefits them, but who hate it anyway, I think the vitriol specific to HFT is that it's a small club and they aren't part of it. Algo-trading jobs can pay significantly better than the best principal engineer jobs at FAANG (which can also be argued to be useless jobs), and those jobs are also much harder to land (think Rentech).
The episodes on compiler optimization, time keeping and multicast are specifically interesting if you want to know more about the oddities of high frequency trading.
> Jane Street Group sued Millennium Management and two former traders over their alleged theft of a highly confidential and “immensely valuable” proprietary trading strategy.
Am I correct in understanding how firms like Jane Street work? They are a market maker - they run an exchange where buyers and sellers can transact. They can arbitrage these trades by connecting buyers and sellers where there is a price discrepancy. Something like that?
Almost always outside of crypto, the market makers and exchanges are different entities. Exchanges maintain order books- who is willing to buy or sell what, at what prices, plus a lot of rules about tie breaking, order visibility, "implied" prices (e.g. sometimes the combination of two products is logically equivalent to a third), etc. When orders "cross"- that is, someone is offering to buy at a price at least as good as someone is willing to sell for, the exchanges matches those participants and they are considered to have traded (though for a mix of technical and regulatory reasons, the trade actually settles two days later)
Market makers generally maintain offers to both buy and sell a product, generally ~all the time the market is open. For example, they might offer to buy up to 30 X for $0.99 or sell up to 70 for $1.01. If small buy and sell orders come in more or less randomly, the market maker will sell about as many X as they buy, for (1.01 - 0.99) a profit of 2c for each set of orders. The trick for a market maker is to offer the best price, so that they get any orders at all, while accounting for the risk that the person buying or selling from them (the liquidity taker) isn't just a random order, but is either market moving or correctly predicting the market is about to move- e.g. a market maker offering to buy a million shares of a X at $0.99 will lose a lot of money to someone who correctly predicted X is about to go to $0.70, and took them up on the full offer.
The big exchanges function much like a traditional exchange and Jane St, Virtu, etc all connect the same way (FIX) to make the market. Really crypto exchanges are just like FX markets.
Very little difference.
They light also be mining, but market making behaves the same as other traditional markets.
The angst against HFTs is odd, as their revenue come from market growth and taking shares away from traditional Wall Street investment banks. Things like ETFs, that most people seem to be positive about, depend on market makers to work. Without HFTs, people just pay more spreads for the same transactions. HFTs are a mirror of the world we live in, not the cause of it.
46 comments
[ 3.6 ms ] story [ 93.4 ms ] threadDoesn't matter if they're so fast they can't lose on any single trade, there's always a risk of shooting themselves in the foot.
So...by capitalizing on tragedy? (I know it's nothing new, it just makes me a little sad)
So they're not adding a lot now relative to the cost and complexity. But banning the whole practice will make the markets function much worse and more expensive for everyone.
Similar for liquidity provision; most of the optimal control theory solutions for market making with alpha signal will widen the spread and remove liquidity the moment they sense something is off.
But if a stock is going up, HFT inserts themselves between the buyer and the seller and takes a small cut in the process. I don't see how it helps the liquidity stock, but seems to be more like an HFT tax.
They didn't increase the number of stock shares, just got in between the buy/ask margin.
Here's a much deeper analysis on bid ask spread from HFT market making: https://www.cftc.gov/sites/default/files/2022-08/HFT_and_mar...
This is a problem of market structures, if I understand it correctly; facilitated largely by Reg NMI in the US and its equivalents globally. And it's not like all was well before that: Instead of HFTs, it was market makers or specialists pocketing impressive, largely risk-free profits.
What you'd ideally want is to preserve the good that the HFTs facilitate (liquidity, more efficient price discovery, smaller spreads etc.) while avoiding the bad (siphoning off investors' money from the markets and math and physics PhDs from the world's top schools), but I haven't yet heard a convincing proposal for that.
Former options trader. This improves dealers’ profits.
Consider a once-a-day auction. The person who puts in an order at 9:35AM incorporates five minutes’ information. The person who does so at 3:55PM, all but five minutes’. Anything material that happened at Noon is extracted from the early order and benefited to the late one. If the orders are unsealed, the later order can incorporate the early order’s existence.
With options, that means the high-speed late order has better information about the underlying stock than the orders preceding it. So it generates risk-free profits from them.
Reality remains real time.
I'd put in lowball bids and ludicrous (but plausibly deniable) offers and wait for the unreasonably-short windows.
The solution to this problem is people not wanting immediacy. (Same with scalpers.) But people like immediacy. Unfortunately, there is also political capital in railing against those who provide that immediacy. In a way, it's a stable socioeconomic system: HFT provides liquidity, market partiipants take it, activists rail against it, regulators complain and win the activists' favour.
https://cdn.cboe.com/resources/membership/US_Options_Auction...
And: https://www.cboe.com/us/options/trading/complex_orders/
Either way I don't think this would resolve the issue with siphoning talent.
It was worse in the "good old days". The cost of trading was much higher and the people that felt that were the people who needed the market to make the world work (risk management, investment etc.)
Markets make the world a better place (it's not a new idea), and more efficient markets are better than less efficient ones.
> If not, how about a per-transaction tax on securities trades, one that wouldn't be out of line for individual investors,
Why do you want my pension fund to spend more of my money on trading costs?
HST turns what would have been a single trade into multiple trades making volume go up and to the right.
> per-transaction tax on securities trades
This already exists anyway. I assume you just want to increase it?
And even disregarding that, practically, the kind of people working in HFT usually have a skill sets that make them highly employable anywhere. If anything, HFT is draining some of the best mathematicians and software engineers from other projects that can't afford to pay as well.
Its more important that we live in huge/new houses and drive new cars and have nice vacations than fund to finding a cure for cancer, and people make this decision every day.
In my view, it's disingenuous and cynical towards the person doing the job: They're effectively historical actors in a museum because we haven't figured out how else to solve for the distribution of monetary and "soft" values like societal status, feeling like a productive member of society etc.
There's nothing at all wrong in my view with letting people do what they want and cross-subsidizing that activity as a society, but why not call it what it is? And why stop at office-style work and exclude e.g. art?
Impacts of HFT on society:
- You no longer have to pay a stockbroker $100 over the phone to take a position and hope they get you a decent price. Trading is instant and spreads are extremely tight.
- Money they make is made by 1. offering you convenience (usually for an imperceptibly small fee built into the spread) or against people trying to do the exact same thing that they are (intraday trading).
The intraday trading is a zero sum game, but the liquidity providing is net positive. The zero sum component only affects you if you're trying to profit from intra-day trading too, in which case, you're a hypocrite for calling them out.
Amongst people who understand HFT well enough to realize it benefits them, but who hate it anyway, I think the vitriol specific to HFT is that it's a small club and they aren't part of it. Algo-trading jobs can pay significantly better than the best principal engineer jobs at FAANG (which can also be argued to be useless jobs), and those jobs are also much harder to land (think Rentech).
The episodes on compiler optimization, time keeping and multicast are specifically interesting if you want to know more about the oddities of high frequency trading.
Here I am not arguing against HFT per se but it is a interesting topic and being critical is a good way to learn.
> Jane Street Group sued Millennium Management and two former traders over their alleged theft of a highly confidential and “immensely valuable” proprietary trading strategy.
Here's the complaint: https://assets.bwbx.io/documents/users/iqjWHBFdfxIU/rEXVjFfB...
Market makers generally maintain offers to both buy and sell a product, generally ~all the time the market is open. For example, they might offer to buy up to 30 X for $0.99 or sell up to 70 for $1.01. If small buy and sell orders come in more or less randomly, the market maker will sell about as many X as they buy, for (1.01 - 0.99) a profit of 2c for each set of orders. The trick for a market maker is to offer the best price, so that they get any orders at all, while accounting for the risk that the person buying or selling from them (the liquidity taker) isn't just a random order, but is either market moving or correctly predicting the market is about to move- e.g. a market maker offering to buy a million shares of a X at $0.99 will lose a lot of money to someone who correctly predicted X is about to go to $0.70, and took them up on the full offer.
The big exchanges function much like a traditional exchange and Jane St, Virtu, etc all connect the same way (FIX) to make the market. Really crypto exchanges are just like FX markets.
Very little difference.
They light also be mining, but market making behaves the same as other traditional markets.