This article (and the comments here on HN) are cringe inducing. The author of the post has no idea how financial markets work. None.
So what if 45% of all trades were odd lots? What was the value of those trades relative to the trades that were on the tape? It's a simple and obvious (and important) question that they don't address in the article because the article is all about pushing some BS agenda that AAPL was somehow being manipulated. Are we really supposed to expect that a bunch of one lot trades were pushing AAPL "to extraordinary highs and confidence-shattering lows"? Are we really supposed to believe that the mom and pop investing public makes trading decisions based on the tiniest orders on the tape?
This article is complete joke and if this is the kind of financial markets reporting you're reading you are grossly misinformed about how the markets actually work.
Totally agree: cringe inducing article. The paragraph that really got me was the following: "Why does this matter? For one thing it means that for years, trading volume in Apple was being misreported -- a situation that can lead to a stock being mis-priced. The system also advantaged one kind of trader over another."
This is wrong on so many levels. Trading volume was not "misreported" it was simply under reported. There is absolutely nothing wrong with that. And it certainly would not lead to any mis-pricings. There is absolutely no way this could advantage one trader over another... none.
There is absolutely no way this could advantage one trader over another... none.
I don't think this is right. Many traders run technical strategies that incorporate trading volume. A trader using the consolidated tape is at a disadvantage relative to another trader running the exact same strategy who uses the full feed.
For example, consider a simple predatory trading algorithm (someone trying to buy ahead of a big player who wants to move lots of shares). If a big player makes a bunch of trades in 99 share lots, they will completely sneak past the predator.
By using the word "predator" you are somehow implying that this is bad behaviour. I disagree with that. Is it somehow wrong to buy because you think the price will go up?
You are further implying that there is 0 market impact from trading odd lots. By "taking" 1 share you are reducing the supply in a symmetric manner. Both demand and supply will adjust to this new information, whether its disseminated to the tape or not.
I intend no value judgement by the use of the word "predator", it's simply a technical term. I'm strongly in favor of predatory trading - it prevents institutional investors from screwing over the little guy by hiding big trades and forcing their counterparties to suffer the price impact.
I'm not implying that there is 0 market impact from trading odd lots. The market will adjust to the new information. It's just that the guy looking at the consolidated feed will not be making a profit by helping it adjust.
I think not printing odd lot trades to the consolidated tape was an anachronism that should have been done away with years ago. I wasn't arguing against that and I think it's a good thing that's happened. My point was that you can't substantially manipulate one of the most liquid, heavily traded stocks on the planet with a few small trades like like the author of the article is claiming. It doesn't work that way.
If there was no advantage to be had, why do computer algorithms include "price discovery" odd lot trades? Honest question, you probably know better than the original reporter does...
Under-reported volume in AAPL cannot advantage one trader over another. I don't disagree, however, that odd-lots can be used by HFTs/algos for price-discovery. I don't think that this necessarily disadvantages any market participants.
Odd lot pings are just ways for HFTs to reduce their capital outlay. You simply don't want to commit too much capital to "ping" passive dark pool orders. So, in the new odd-lot reporting method, oddlot pings will now be reported to the tape. I would argue, however, that this makes no difference to the HFT. The consolidated tape is so slow, that by the time other HFT participants can digest an odd-lot fill from a darkpool, that remainder of the passive dark order would have already been gone/hit/lifted by the pinger. I view this behaviour purely as proper functioning of the markets...
Can you elaborate on the difference between "misreport" (def: a false or incorrect report) and an "under report" (def: a report that reports a number below the truth)?
The effect of under reporting was due to the rules and regulations around how trades are reported to the consolidated tape. Those rules changed recently to bring the consolidated tape volume closer to the actual volumes. Misreporting implies malicious intent.
I don't think the term "misreport" implies malice; just that the report be inaccurate.
A temperature sensor that both reports both high and low temperatures is said to misreport the temperature.
I understand your comment, however, if you are saying that it wasn't malicious - it was misreports stemming from decisions about how to implement the rules.
The conclusion wasn't that this was caused by HFT but because NASDAQ and the NYSE only showed trades that had a size of 100 or greater in their public feeds. Since a 100 shares of APPL is worth $50,000+, there's a lot of trades being made for less than 100 shares and, until the reporting changes made a few weeks ago, they didn't show up on the public feed. This means that trade volume wasn't being reported accurately, and that the public feed wasn't representative of the active price of a stock like APPL - which is a fairly big deal when, according to the article, 45% of APPL trades are in lot sizes of under 100 shares.
For professional traders (and here's the only place where HFT is relevant) it's not a big deal since they don't use the public feeds but instead pay for feeds from the exchanges which include all trades.
It's a big deal to everyone who is not a "professional" trader. Oh and by the way, HFT mostly trade low volume at high speed. In other words, trading 5 shares of Apple 5 million times an hour. That makes money. It's not about the big pay-off. It's about the pennies all adding up. And also explain why the stock was wildly moving all over the place.
Personally I think this type of trading should be illegal. Right now, unless you have a data center right next to the exchange you are trading at a disadvantage. And it just adds to the many reason why I think the game is rigged. To favor the big houses. You're almost better off at Vegas.
In my experience HFT is a boogeyman. It's not well understood, and it's easy to blame for ills real and imagined - in part because you only tend to hear about it when something goes wrong and a Knight Capital type event happens, although that can apply to any firm that does any form of automated trading such as Market Makers and Algo traders and is definitely not the sole preserve of HFT shops. Based on what I know and see, I believe that the existence of HFT is generally an aid to price discovery (due to the narrower spreads), and liquidity (more actors means a better chance that someone will take the other side of a trade). In that respect, they're not too dissimilar to Market Makers, although a true market maker has additional responsibilities in terms of being willing to trade in illiquid products.
If you're not a professional trader then the topic of the article really isn't a big deal. If your trading style is such that it's significantly material whether you trade at 510 or 511, or whether your trade executes in the next minute or the next thirty minutes, then you're always going to be at a huge disadvantage compared to people who can get the best price and most immediate execution. A trading firm who are willing to pay to connect directly to an exchange will always get a better price than you can.
It's not horribly different to why the restaurant down the block from you gets better prices for produce than you do - their business affords them better access to the produce market.
Edit - For what it's worth, the bulk of my career has been in finance, mostly as a Quant-Developer at a mix of buy-side and sell-side firms and more recently as technology consultant who specializes in Capital Markets.
Yes, the never ending and hard to pin down boogeyman. Today it's HFT, tomorrow it's day traders, the next private equity. All these thing are so evil, so easy to do, and the people who do them are making so much money, which should be our money really because they're robbing our 401Ks blind.
Of course, I'm being sarcastic, but this is what media is selling and most people are buying it without any pause for reflection.
There is an obvious problem with HFT: it gives an advantage to lower latency analysis and communication, leading to huge amounts of wasted effort getting faster pipes and software, co-locating automated trading, etc. None of this is delivering real value to the economy, as far as I can see. If you have an argument for why trades that happen at millisecond time scales help liquidity and price discovery over what you'd get with a system that handled trades on the scale of minutes, I'd like to hear it. I have a hard time imagining that the benefits (to society at large) could outweigh the costs of the waste when I see what's being done.
That said, I agree that the high frequency of trades themselves are probably not the problem. The problem is rewarding a latency/processing advantage.
If a HFT firm makes a profit then they are, by definition, adding value to the economy.
All that technology investment is not a waste; it allows firms to condense money out of information asymmetries with ever greater efficiency.
Edit to add: if you prefer industries that build "real things" then look at the revenue that HFT delivers to companies who make high-speed networking and computer equipment.
My post was comparing the current system that allows millisecond scale trading with a hypothetical one that slows trades down to something on the order of minutes. Your observation that people are willing to pay to exploit information asymmetries in the current framework doesn't address the question of how my proposed change would affect net economic value.
The metric of measurement is wealth, which profits create.
Your proposed change would reduce the benefit because it would eliminate the profit (wealth) that HFT firms create.
Your proposed system is not hypothetical; it is how all trades used to happen. HFT did not arise from nothing. People developed it because it created new profit in addition to that generated by slower trading.
Suppose I'm a speculator willing to pay $0.03/share for liquidity. It's valuable for someone (I don't care who) to provide that liquidity.
If HFTs were all 100ms slower, I wouldn't care. But if HFT 1 is 100ms slower than HFT 2, then HFT2 will capture $0.03/share and HFT 1 captures nothing. It's a wasteful arms race. It's also fixable simply by eliminating the subpenny rule.
It's not that it would eliminate it, but it would make the margins to due pure latency arbitrage so small as to make it not worthwhile (in theory anyway).
That said, there is and always will be an advantage to players that can respond more nimbly to changing market conditions (news being one input to this) and there is no inherent problem with that (and there may be lots of positives to the market as a whole).
Have you considered this post in light of the inverted exchanges BYX, EDGA, and BX? You can effectively make your liquidity cheaper by posting to an exchange with an inverted fee schedule (pay to add, rebate when you take), but we still see the "latency arms race" at the traditional exchanges.
Moving the decimal point to the right just increases by 10x the volume necessary to create the same profit. I think a likely result would be massive consolidation to a couple big companies, who would then compete once again on latency but at much higher volumes.
If you're investing over anything approaching the short term then it is exactly like playing at Vegas, except the card counters aren't banned.
If you invest, i.e. over 10+ years then HFT doesn't matter other than giving you better liquidity and lower spreads. There is much, much less competition in that space, no computer trading and you are capable of actually doing research.
Investing and watching daily stock moves is a sure fire way of losing your hair and becoming miserable.
Individual investors tend to outperform the market with a 20 day holding period, with "stocks experience statistically significant excess returns of 0.80% in the 20 days following a week of intense buying by individuals, and −0.33% following a week of intense individual selling" [1]. The individual excess returns are attributable to the "contrarian tendency of individuals leads them to act as liquidity providers to institutions that require immediacy," as evidenced by "the magnitude of the excess returns" being "greater in less liquid stocks."
Shorter than 20 days and individuals compete unfavourably with professional traders' infrastructure. Longer and they compete unfavourably with the skill of professional asset managers.
The number of professional asset managers who actually invest for periods exceeding a year is surprisingly tiny. The number of firms (especially American) that wouldn't fire someone who didn't consistently meet/beat expectations each quarter is low.
I work for a proper long term investing asset manager and we are very happy with the lack of competition in our space.
Short term "investors" are really speculating, which is fine, but call it what it is. There's also nothing wrong with using professional managers to look after your money as well, picking individual stocks takes time and paying someone else to do that is perfectly sensible. Just pick someone who genuinely invests for the time period you want to invest with them for.
>If you invest, i.e. over 10+ years then HFT doesn't matter //
Is there any proof of this. How long have we had this sort of HFT?
Strikes me that HFT is extracting value from the market. The prices are usually relatively stable they're increasing the draw of money away from smaller traders but also removing value overall (in the form of money) from the system and not adding much value. Yes, liquidity, yadda-yadda but surely market liquidity is no greater now but HFT is far more effectively removing money (a proxy for the value extracted).
In other words other people are creating value and HFT is a sink for it. The bar to entry to the HFT camp means there should be an overall movement of capital towards the most wealthy in this sort of system.
HFT extracts the value from transactions. It won't extract the value from a company becoming twice as valuable.
Invest because you think the company is going to be successful and thus more valuable. The actual amount of stock that is HFT traded is tiny relative for each company, long term price is driven by what institutional investors are willing to pay for actual percentages of a company.
Why are the other people creating value, while HFT traders do not? Both types of players extract value from the market, while improving market efficiency.
HFT definitely has high entry costs (10's of millions of dollars), and firms that aren't serious about HFT can't participate. Note that HFT only makes 2 billion a year, so they aren't actually missing out on that much. HFT just isn't a very big space, compared to long term investing, and doesn't represent very much movement of capital. (The wealthy don't really invest in HFT, since HFT firms generally don't accept outside capital, or need any substantial amount of capital to operate)
>Strikes me that HFT is extracting value from the market. The prices are usually relatively stable they're increasing the draw of money away from smaller traders but also removing value overall (in the form of money) from the system and not adding much value.
I'm fine with and agree with that, but my fear is that the transition of the exchanges to public companies (which in turn drove the embracing of HFT) and proliferation of HFT is skewing the market one way or the other.
When you have an environment when 45% of the trading activity for a heavily traded stock is obscured for many years, I don't have the ability as a layman to figure out what that means to me and the pricing of the equity.
Many folks would respond "than you shouldn't be investing in individual stocks", but still, if I'm investing in large-cap ETFs, I own plenty of Apple shares as well as other companies that may be affected by this information asymmetry.
When you have an environment when 45% of the trading activity for a heavily traded stock is obscured for many years
I'm not sure what you mean by "trading activity," but this article says 45% of the trades, _not_ 45% of the volume. We don't know what fraction of volume because the author doesn't bother to tell us. That's why the article is crap. It makes a bunch of extremely contentious claims and doesn't answer any of the important questions.
Many people are in company mandated 401ks (full of conflicts of interest between the trustee and the trustee's own funds) that don't have the option to buy and hold a single stock for 10 years.
Generally, HFT firms end up decreasing volatility. They price stock more accurately, which makes them money and decreases volatility. If they were increasing volatility, they would make infinite money, since the source of their profit is the volatility.
The main scenario where HFT firms increase volatility is when their algorithms break (either due to a programming error, or because they cease to accurately model the market due to a change in market conditions). In that case, the misfiring algorithm will increase the market volatility, and lose the HFT firm money.
You absolutely right that firms without co-located data centers are at a disadvantage. Thankfully, this only really matters if you are trying to compete with HFT firms. So it's not a big issue for people who trade on a longer time scales. The HFT firms do make some money off of these long term investors, and in exchange provide increased liquidity, decreased spreads, and decreased volatility. (if the HFT firms weren't offering the best prices for the stocks, then the long term investors wouldn't be buying from them)
Great, could you explain to me as a retail investor without direct access to the exchanges how to place an ALO order and to buy or sell without enriching HFT shops? (Hint: you can't.)
With a retail broker it's moot - they charge you a flat $8/trade regardless of whether you add or remove liquidity. You never collect rebates or pay to take liquidity.
You can still choose not to cross the spread with a retail broker. Just place limit orders on the right side of the spread.
Most retail traders actually get slightly (insignificantly) better prices for liquidity taking orders through price improvement at the wholesaler. Retail traders can also send post-only orders and participate in exchange the rebate/fee program on Interactive Brokers.
Hft market makers, the ones that thrive on collecting fees for for trading, are required to send orders in even lots. They can only send odd lots to hedge, and even that is a recent development.
Lower volume customers have lower fixed costs, they generally get much worse quotes (higher spreads) too. The fact is whilst $56k might sound like a lot to some person day trader, it is nothing compared to the actual market cap of the stock.
It also has absolutely nothing to do with the use of HFT. HFT only competes against other HFT. There is nothing here to suggest it has been effecting the price for the small customers (normally this is what HFT is credited at doing well, providing liquidity and removing pricing inefficencies) People hate HFT because it can go 'screwy'.
Every trade will affect the price of a share but that's not manipulation, that's a functioning market. Unless you can explain why a trade made by HFT is somehow more devious or underhanded than a trade made by another actor in the market.
How? Can you explain the specific mechanism (i.e., what orders are placed by which party and when) and the role that "high frequency" plays in the mechanism?
That isn't really an example of market manipulation. Market manipulation is an intentional manipulation of the market. (http://en.wikipedia.org/wiki/Market_manipulation) The Knight Capital incident was an example of a firm performing bad trades due to an error in their systems. Their behavior certainly wasn't intentional or profitable.
HFT isn't based on market manipulation. Some HFT firms perform market manipulation, as do some non HFT firms, but this practice is illegal. (For example, newedge was recently fined for market manipulation)
It can even happen to manual traders. If you accidentally screw up and run the wrong strategy (e.g. buying Tweeter Inc. when you intended to buy Twitter), you'll very rapidly lose gigantic piles of money.
Losing your money in the stock market != market manipulation.
I find this sad. The article doesn't blame HFT in anyway and yet the top comment is an "add nothing to the discussion, just give me karma" throw away comment.
Hacker news seems to be doing alot of knee jerk reaction voting here.
What exactly about the article gave you "another sign" that HFT is a bad idea?
There are certainly things that HFT firms do that aren't great, but this article doesn't point out any.
I would have backed the comment up further with more comments, but I went to bed. Beyond that, I dont like HFT because I believe it increases volatility, and even if I'm wrong on that point it's also not something the average Joe can participate in, money or not.
Not being something the average Joe can participate in isn't by definition bad. There are lots of things the average Joe cannot participate in (for example, charging others money in exchange for legal advice, or performing surgery) that I assume you are in favor of. In some cases it's not optimal (requiring bar passage and a law degree to be a lawyer is arguably just a method to restrict competition in an area where all of the necessary material is publicly available, plus law school notoriously does not prepare you to be a lawyer), in others it is (the consequences of an improperly trained surgeon, or pilot are disastrous).
Replace HFT with money and you're on the right track. The bad behavior in terms of investing will always exist because of greed. Greed will always exist as long as humans exist.
So what if 45% of all trades were odd lots? What was the value of those trades relative to the trades that were on the tape? It's a simple and obvious (and important) question that they don't address in the article because the article is all about pushing some BS agenda that AAPL was somehow being manipulated. Are we really supposed to expect that a bunch of one lot trades were pushing AAPL "to extraordinary highs and confidence-shattering lows"? Are we really supposed to believe that the mom and pop investing public makes trading decisions based on the tiniest orders on the tape?
This article is complete joke and if this is the kind of financial markets reporting you're reading you are grossly misinformed about how the markets actually work.