This is a pretty well researched article but the conclusions are rubbish. As a person who has worked for all parties involved in the HFT space i.e. both buy side, sell side and the exchange I can safely say that this is rubbish…you should put your thinking cap on and wonder whats the motivation behind exchanges spouting such rubbish…maybe technology investment is something to think about? These behemoths like CME, ICAP etc…are lagging years behind the market. Ask yourself what their role is? To facilitate an efficient market place… that’s what they get paid for. They have failed on all… Read more »

I think the worst combination is HFT with last look privileges. Hotspot cleaned house and threw out 15 of their 35 LPs late last year, apparently customers complained too much about rejections. I mean, being able to sit on an order for some time and then decide to fill it or not sounds like the wet dream of every hft – but it seems like customers/takers are increasingly prepared not to accept this stuff.

Overall this is a balanced article. You should clearly differentiate between the major styles of venue when discussing. For example there’s no obligation to market make in the otc arena (EBS/Reuters). There may however be a contractual role here between the broker and the trading entity, with attractive brokerage arrangements. The other end of the spectrum has the single dealer platforms, where there is an obligation to market make, and a client/LP relationship to maintain. This relationship manages the margin impact of the HFT player on the LP. IMHO the innovation in the market data, to provide DOB data, can… Read more »

This is a very interesting article, well reserached. Let me add a few comments from a perspective of a buy side person working with one of the top names in ECN space. Providing liquidity over ECNs has become a very competitve thing, I am sometimes surprised at how narrowly top LPs quote, at least for typical retail clients. 0.4-0.5 pips doesn’t really come up to “lucrative margins” so I disagree with the author here. most LPs are able to generate 10 or 15 bucks per 1M traded after bro and not all retail brokers generate 1B ADV. Given a smaller… Read more »

Low latency access permits more fairness in the market. Instead of using unfair rights such as Last Look, LP who want to protect their back needs to quote well and fast. What is the problem with that for the market participants?

The true problem is the bigger player who wants to keep their birth rights.

The whole industry fights HFT to protect the small players… What’s a joke…

Last look should be banished. Low latency access developed. And market abuses, HFT or not, punished. Tools are now available for that.

Max: in futures and equities they never had LL, i don’t see a compelling reason why it should not be possible in fx..

Ron – very well written and informative article. And the comments generated make this one of the best threads in recent ForexMagnates history. Nice job all.

@Mark. Good point that many miss is that the ECN’s each have hundreds or thousands of brokers/MM using their platforms….but relatively few LP’s pricing into the ECN’s. Therefore the ECN’s are very interested in keeping the LP’s happy. Losing one broker/MM is bad, but losing an LP is worse.

The new ECNS are interesting as some are anti HFT and some pro.

@Andy: I think you overestimate the effect of last look. I am in the retail business and over 99% of the flow is what you can call benign friendly flow. What really counts to my clients in terms of transacting is: narrow spreads (what the see), short response time and low slippage. Having relationships with LPs I can influcense all three factors and I can literraly control the last 2 factors. My thesis is that last look is a fee that I pay to get a spread reduction. Lets say that each one of my trades that was last looked… Read more »

Perhaps there could be a master list of PRO HFT venues and ANTI HFT venues? This would make it easier for a trader to make a decision

Often times a retail broker will say “all EAs/trading strategies welcome” and then the trader proceeds to trade only to find out sooner or later, slippage, artificial delays, and other dealer intervention is taking place at least opportune times.

I totally agree with commentator dj: EBS and CME HAVE LOST the technological advantage that they used to have. EBS is such a boring company that I fall asleep when talking to their Key Account Managers (and I had a couple of invitations to join their club). Now, they are behaving like little cry-babies including their masters – some (not all) of the Tier1 “liquidity providers. Yada,yada, yada. It´s a shame they are losing business to the more sophisticated outlets. Apart from this observation, one forgets only too often that the main participants in markets are CENTRAL BANKS. As a… Read more »

@Ron: London Stock Exchange made a couple of deals with Fixnetix, GateLab and Millenium IT (a joint venture with Intel). Yet, I don´t know whether this technology is being used by LIFFE. Also, exchange connectivity is not the only criteria to decide for CME/LIFFE. It´s also about issues like pre- trade risk management procedures and some other regulatory burdens.