Hacker News new | past | comments | ask | show | jobs | submit login

Having read 'Flash Boys' by Michael Lewis, this post seems like the logical followup after someone discovered the first microwave tower at the end of the book.

Didn't know that the same scam is running in Europe...




HFT is not the same as front running, this post does a good job in explaining why. http://www.chrisstucchio.com/blog/2014/mark_cubans_hft_idioc...

There are other posts by the same author which cover the same topic if you're interested.


High Frequency Trading should have been called Low Latency Trading.


There is a real problem when discussing this topic that many terms are bantered about without rigorously defining them. Most electronic trading (trading done via computers and computer networks instead of in person) exists on at least 4 spectrums.

1) How algorithmic is it? How much of the trading decision is made by a computer and how much human intervention is there.

2) How often does it trade? Some electronic trading systems can trade thousands and thousands of times a day, while others only trade a few times a quarter.

3) How long does it hold a position and/or leave a quote on the market? Some trading systems are designed to never hold positions and others hold positions constantly.

4) What latency requirements does the trade have from the time the exchange publishes data until the time an order needs to be in the market.

There are electronic trading systems that exist with variables at every end of these spectrums.


It was at first. But it didn't stick.


How is microwave dish based communication a 'scam'?


It introduces asymmetry of information. Some forms of such asymmetry are obviously unethical and illegal - the good old insider trading. Some are legal and ethical(#) - just having better research on public information is ok.

Buying extra {milli, micro}seconds to get ahead of the competition is just too new to have a general consensus on its ethical status.

(#) In the mainstream. Obviously you can find contrarians to any position.


A) these microwave towers are mainly on routes to/from commodities exchanges that have equities derivatives and the equities exchanges where the underlying equities are traded, or between 2 different commodities exchanges. Insider trading, you will find, is much less regulated in commodities trading as there is all manner of information asymmetry built into the market, in fact one of the purposes of the market is to discover a valid price in the face of this information asymmetry.

B) Buying extra time via technology is as old as trading. If you read the article, or taken an architectural tour of the CBOT building in Chicago you will have noticed that one of its main design requirements was to house telecommunications gear.


Only because B) has been there for ages doesn't mean, it has to stay that way, does it?


I was specifically replying to the idea that we haven't had time to judge the ethics of communication technology in trading, when it is in fact a basic market force.

I'm not an ethicist so I'm hesitant to speak on the ethics of market forces, but I will say:

A) How do you prevent some actors from getting information before others? Do all actors have to wait for the slowest participant? Why should we prevent it?

B) In general, faster dissemination of information leads to a more efficient market, we should encourage this, not discourage it.


People have been using technology to discover news sooner for centuries. Your ethical compass is too slow!


The scam is what you use the microwave communication for. If you have access to fast lines or even faster radio comms means, that you can rig the market more efficiently.

It's called front running.

http://en.wikipedia.org/wiki/Flash_Boys


The book has been discussed here at length[1]. I never worked in finance but it doesn't seem to hold up very well under scrutiny from experts. Comments by 'kasey_junk and 'yummyfajitas were particularly informative. The latter even wrote a couple of blog posts, including "A Fervent Defense of Front-running HFTs"[2].

[1] https://news.ycombinator.com/item?id=7531429

[2] http://www.chrisstucchio.com/blog/2014/fervent_defense_of_fr...


Being faster is rigging the market? How does that work?

Also, it isn't called front running, at least not legally. It can't be, HFT don't have any customers. Maybe Michael Lewis has a different definition of front running but for the rest of us: words have a meaning.

http://en.wikipedia.org/wiki/Front_running


HFT might not have direct customers per se, but if an order matches, they have a counterpart on each trade they make. If you have an unfair advantage for that trade, what would you call that system? Rigged? That's what the big players in the HFT field are doing by using the various tools available (near-location, co-location, fast networks, fast custom hardware and lately microwave networks).

On the term, you're right - front running might not have been the correct therm.


Every trade that happens in every market in the world can be viewed from a game theory perspective as adversarial. In the absence of other outside factors each trade in isolation is a zero sum game, that is 1 counter party will make money on the trade while another loses (at least in the form of opportunity cost). In this adversarial encounter it is expected and encouraged for all counter parties to strive for unfair advantages. Some do this by using vast swaths of overworked and underpaid ivy league grads to do market research for them. Some do this by using massive amounts of leverage others do it by being fast, some use illegally obtained proprietary information and still others do it via mobbed up boiler rooms. Some of that unfair advantage we have deemed illegal and others we have deemed legal. The point is, the existence of an advantage in a trade is a necessary condition for the market to operate correctly, not evidence of it being rigged.

You will notice that the "big" players in HFT are much smaller than the people they are taking advantage of, the giant hedge funds. The HFT competitors have access to the same technology, technologists, exchange access and microwave networks. Further, they have better access to capital and legislative power, not to mention access to best selling authors.


> Rigged? That's what the big players in the HFT field are doing by using the various tools available (near-location, co-location, fast networks, fast custom hardware and lately microwave networks).

I can't personally afford to have a Bloomberg feed in my living room to trade off of. Why is this any different?


Neither can I. Do you feel having the same fair market access without the BB terminal at your disposal?


Does it upset you when you go to buy a car to think that Avis can buy that car at a gigantic discount because they buy in bulk?


But it does if I have to buy it at higher price just because Avis knew a that I and a few fellows where to buy the car 2 min in advance because he saw us asking about that car, bought the available stock of car we were interested in to the car dealer, propose them at a slight premium to the the car dealer who then finally sell it to me and my fellow 1 minute later at another premium because, you know, it has to make its own margin. The whole thing happening just under my nose.

That's how I (unobjectively I concede) feel about HFT.


Ok, now let's take that thought experiment and make it closer to the real world:

Imagine that Avis is buying cars in bulk and getting better deals as a result, and then imagine that the knock-on effect of them doing that is that you get better deals on cars because of how they change the way the car market works.

Car dealerships are, for obvious reasons, really fucking upset about this. And they are noisy and well-funded. Normal people start getting angry at the giant rental car companies and the unfair advantage they get, unaware of the fact that without those companies, they'd be getting screwed even worse by the dealerships.

That's what HFT does to normal investors. You don't have to take my word for it; the Chief Investment Officer at Vanguard, among many other people, has said exactly this.


If you create an ad on television stating that you are buying tons of a particular model and make of car, and then demand for that car goes up, why is this seen as a negative in a functioning economy? A functioning economy responds to demand.

Now what is the difference between sending an exchange an order for more shares than they have available and a television ad broadcasting your buying intent? In this day and age: not much.

What isn't happening: small traders, like you and I, are not sending small trades to markets and having them be "intercepted" by HFTs. HFT's wouldn't pay attention to the "guy buying one car" type of trader, and there isn't enough of a difference in the signal (your trade) and the current market price to be arbitrageable.


This would just be price correction, if a given car is becoming popular then it makes sense for it to cost more. Likewise if a given car is becoming less popular Avis would be willing to shift their excess stock at a discount.


I like where this is going :) I know, I have a choice of buying a car or participating with my own money on the stock market.

I have no choice if the pension fund where I work at is investing in blue chips.


Once again, this time with feeling: HFT firms don't prey on pension funds, mutual funds, or value investors. They improve things for those kinds of investors. HFT firms prey on other sell-side firms.


Front running is something else, http://en.wikipedia.org/wiki/Front_running


"scam"? Care to elaborate?

HFT has been around in Europe and is as well established as the US, so I assume the alleged scam relates directly to buying and operating Microwave networks. Seriously, please do provide more information, I am keen to learn more.

disclaimer: I've not read 'Flash Boys' yet, so apologies if it's all explained in that book


If you are interested in the topic of HFT, 'Flash Boys' might literally be the worst introduction to the industry you could read. It is rife with bias and technical incorrectness. Worse than that it is disjointed and plain poorly written (and a fair chunk of it appeared in Vanity Fair which is available online).

I have my own problems with 'Dark Pools' by Patterson, it does have some technical inaccuracy, but on the whole it paints a picture that is closer to reality and is still readable.


Any other suggestions for reading materials on this? Sounds like slightly faint praise for Dark Pools.


If what you are looking for is "narrative" non-fiction, Flash Boys and Dark Pools sort of own the corner. Of the two, Dark Pools is much better. In general, it is hard to tell a good narrative around HFT because it is so technical.

Trading & Exchanges by Harris is dated but will give you an introduction into market microstructure, keep in mind it was written before RegNMS.

If you want a good set of blog posts, Chris Stucchio (yummyfajitas) has an intro series that is great on his blog: http://www.chrisstucchio.com/blog/2012/hft_apology.html


Reading TCP/IP Illustrated won't teach you how to configure BGP prefix filtering regexes on a Cisco router, but the background sure is helpful. Harris is similarly helpful (and in a bunch of weird ways, similarly constructed and written) for understanding financial technology. I do software security consulting for large financial technology companies, and if you ever want to get one of their developers head's nodding, mention an example from Harris.

It's weird that more nerds on HN haven't read it. It's a great book, and it's great in a very nerdy, systems-y way.



Read the book and you'll learn why it's a scam. Only because something is well established, doesn't mean it's morally legal. Your pension funds (and so is mine) are getting screwed big time by HFT.


No they aren't. One of the largest and hands-down most respected mutual fund companies in the world, Vanguard, the company that inspired the "Bogleheads" movement, publicly stated last year that HFT had helped them, by driving down spreads.

The fact is that credible pension funds don't make their money by competing with sell-side firms. When your only job is sporadically making large block trades, you're happy to see spreads competed down as far as they can go. You don't care which market makers are winning the race to provide the cheapest possible liquidity.

Claims that buy-and-hold funds are victimized by HFT are hard to square with what Vanguard said, and with the logic of how automated trading works.


Your pension fund might be aggressively speculating in daytrading. Mine is not. You might want to look into that for reasons beyond other market participants. You can lose your shirt if you're a short term speculator very quickly, regardless of presence or absence of HFTs.

Long term investment might be a better strategy than wild speculation for a goal like that.


So what are pension funds doing about it? If they're getting screwed, they should be speaking out. Are they?

EDIT: Replying to my own question: http://blogs.wsj.com/law/2014/09/08/law-firms-sue-stock-exch...


I believe in Flash Boys the issue is not HFT itself but that the way the market provides access favors certain parties. I could be wrong though, I've only seen the (somewhat related)documentary on the subject some time ago.


Yes, some are, some will. And there's a new exchange open with fair access: IEX.

http://www.iextrading.com/

http://en.wikipedia.org/wiki/IEX

"IEX's main innovation is a 38-mile coil of optical fiber placed in front of its trading engine, which adds a round-trip delay of 700 microseconds and is believed to limit traders' ability to respond on the dark pool ahead of IEX's own pricing algorithms."


IEX is not an exchange. It is a dark pool with technology, rules and special order types specifically in place to give large hedge funds advantages over "average" investors.

They still allow collocation, venue arbitrage, and non-human timescale trading. Any electronic trade you could do on any other dark pool you can do on IEX.

The main difference between IEX and other dark pools is that they are owned by hedge funds and they either duped or bribed a famous author to write about them.

[edit] The comment below correctly points out my insinuation about bribery and Michael Lewis is unfair and snarky. I will point out that Michael Lewis wrote a glowing book about Jim Clark a decade ago, who happens to be an investor in IEX, so there is at least the potential for intentional bias.


"they either duped or bribed a famous author to write about them" is that just your personal opinion or do you have any evidence to back this up?


Let me save us all a giant subthread:

https://news.ycombinator.com/item?id=7531429


All of that is just marketing talk.

IEX is currently not an exchange. What they offer is a dark pool connected in series with a Smart Order Router. That's what all the big brokers (e.g., JPMorgan, Morgan Stanley, Goldman, Deutsche Bank, etc.) have been offering to their customers for years and their dark pools usually offer a lot more liquidity. The big brokers have become better and better and smart order router technology to protect their customers as much as possible from HFTs. IEX is at least 5 years late to the party.

Perhaps they are planning on becoming an exchange one day, but I am not sure how they can do that while slowing down order executions. It seems to me that that would be a serious violation of Regulation NMS.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: