I wrote a while ago about the impact on fragmentation caused by an outage at NYSE Euronext (see The Croissant Hypothesis, April 2009). Well, we had another chance to re-run the experiment here in London this week. Basically, what happened was that on Wednesday the LSE suspended trading in a small handful of stocks because of some technical glitch in part of its market data feed. Spotting the opportunity, all the MTFs enthusiastically emailed the market confirming that they were still open for trading in the stocks in question. But, just as happened in Paris back in the spring, the expected surge in fragmentation failed to materialise and London traders simply munched on their bacon rolls until the LSE fixed the problem a few hours later.
This was shown by the fact that the average FFI of the stocks in question (2.25) was, in fact, slightly lower on Wednesday than it had been for the previous week (2.3).
This highlights, yet again, the subtle interplay between the MTFs and the primary markets. There appear to be three reasons for this. Firstly, the primaries are responsible for maintaining an orderly market in the stocks they list and so traders are reluctant to trade a stock without its “parent exchange” being open. Secondly, a lot of liquidity providers use the primary market price as the starting point for their calculations – without this they simply can’t make markets in those stocks. Finally, there is the psychology of the market that still feels that stocks “belong” to particular markets.
So, the paradox of the current situation is that without the primaries the MTFs simply couldn’t exist. The bizarre implication of this, then, is that the LSE could beat the MTFs simply by taking its ball home and refusing to come out to play at all but, that way, everyone loses. More likely is that the calls for a consolidated tape will be re-ignited yet again but, in order to really fix the problem, any such tape would need to be fully mandated by the industry. My suggestion is that we all get together to define this and fix the problem – otherwise we may find the regulators doing it for us.
If you would like to be part of such an initiative then please let me know.
Just to concur with todays blog, I do believe that the MTF volume is “artificial” in the sense that you have a lot of MM prices generated in-house or at the MTF servers, by liq providers. This is quite evident both in the topic you raise in the article, but also the somewhat desperate search for client flow which seems to be in scarce supply. Are we going to have the same development in Europe as in US, i.e. the large orderflow providers opting out of exchanges and making private arrangements with exclusivity to some banks who guarantee liquidity ? Time will tell…
Will it be possible to see the Fragulator on an index eg FTSE100 for yesterday?
Thanks for your query, NS, it’s one that a number of people have mentioned. Yes we do plan to add indices to the Fragulator in the same way that they are covered on the main FFI website. We also plan to include dark pool breakdown so you can see the total market share that say Liquidnet has over a particular time period.
I am interested (only) in ETFs. Could you create something that shows ETF fragulation?
Steve
In fact, at least one industry practitioner called on Fidessa to lead the consolidated tape charge at the seminar you guys organized not so long ago, you may recall.
Carl – that’s a great idea and one that a number of people have asked about. We certainly plan to include ETFs because of their growing popularity. It would also provide another level of contrast in-between fragulating at a macro Index level or at the micro single stock level.