What gave you the idea?
We were trying to get a picture of what was going on. But each MTF was calculating its market share differently, so it was very difficult to get any kind of meaningful comparison. The second thing was, it was very hard to get an idea of any trends. So we gave the R&D people the challenge of finding a way to measure how fragmentation is occurring, so we could predict trends. And they came up with a calculation based on the inverse of the Herfindahl index, which is widely used to measure concentration within an industry sector) . [The maths is at http://fragmentation.fidessa.com/ffi/.]
And how is the index being used?
What's useful is that you have one number that's telling you, on average, how many venues you would have to visit in order to complete a trade in a particular stock. The higher the number, the more venues you would have to visit.
Obviously, it's a useful reference point for market commentators, but apart from that, who else is using it?
First, the broking community need to understand how much investment they should be making in smart order routing technology. By looking at the stocks they trade, and monitoring the index over time, they can form useful views as to their smart order routing strategy. Secondly, on the buy side, a firm with multiple broker relationships can look at the FFI for their portfolio of stocks, and get an indication of how they could route trades. A stock with a low index number could be traded through a low-cost DMA broker. A high number would indicate a stock worth trading through a broker with smart order routing capability.
What trends in fragmentation would you point to today?
There are a few things. You can see, for example, that the FT-SE 250 is a lot less fragmented than the FT-SE 100. Also, you can see that the rate of fragmentation in mainland Europe, if you look at the CAC-40 and the DAX, seems to be increasing.
Also, it's early days yet, but you also get a sense of who looks to be getting market share among the MTF community. Turquoise is doing healthy volumes in some stocks, for example, although not at the expense of Chi-X. That's new volume they're getting. BATS has made a healthy start, aiming itself very squarely at the high-velocity traders.
What's really interesting is that it's such a dynamic market. For example, this is the year in which the traditional exchanges fight back. They've got natural advantages, in terms of brand, distribution, reach, so it's by no means a certainty what's going to happen.
Do you think fragmentation's here to stay?
It's impossible to put the toothpaste back into the tube. Fragmentation is definitely here to stay. The exact shape of it is undecided, though. This year, there are a lot of dark pools being launched, and exactly how the trading interaction works between dark and lit venues is going to affect who the winners are.
Fragmentation is here to stay, yes, but it's by no means guaranteed that the winners will be the MTF community. The general consensus is that volumes are going to be lower this year than last, so to hit their break-even targets, the MTFs are going to need an even bigger slice of the cake.
What about consolidation?
The trend we see is, we're going to get a period of increased
fragmentation. There are more MTFs to launch. In the end, there
will be consolidation, but perhaps along different lines, on a
sector basis for example rather than into national boundaries.
Do you see fragmentation as a good thing? A benefit to the European market?
While fragmentation has driven trading costs down, you've got to add back in the cost of smart order routing, and the added complexity of clearing and settlement. Because clearing and settlement regimes are also fragmented, I'm not sure that MiFID's objectives are being achieved.
Could the FFI become a tradable index?
One of the things we're starting to look at now is the volatility of fragmentation. One thing we can see is that the volatility of fragmentation of some stocks swings wildly from day to day. We're not sure what the reasons for that are. A volatility index based around fragmentation would take some of the risk out of that.
Are you planning any more indices?
We have one or two ideas. Latency, for example, is an increasingly important area for people. On the near horizon, we're looking at how we measure all the dark pools and how we treat them in relation to the traditional exchanges and the lit MTFs.
Time for grown-up thinking
At the end of 2008, fragmentation became the subject of a 'Vox Pop' at www.automatedtrader.net. In a notable entry, Dave Lauer, senior systems engineer at Tervela, gave his view of what's happening, what needs to happen and what might happen.
The liquidity fragmentation trends in Europe have substantially accelerated since MiFID regulations have begun to be enforced. This fragmentation is leading to a landscape very similar to the highly-fragmented US markets, primarily concentrated in traditional venues with an increasing amount of liquidity to be found in dark pools and other less transparent execution venues. Though some would argue that requiring best execution protects against liquidity fragmentation, latency in market data and computational impedance in the order flow process actually exacerbate it.
While there are significant differences between MiFID in Europe and Reg NMS in the US, both market landscapes are dramatically changing. With a projected CAGR growth rate of 46% over the next two years for the percentage of value traded by off-exchange crossing networks, anyone working in the industry - buy-side, sell-side and vendors - must pay attention to current trends.
The advent of fragmented markets is an example of pure competition - many venues using innovative technology, price improvement strategies and pricing incentives to meet differing requirements. While this sounds wonderful in theory, the current reality became all too apparent in September 2008 with the LSE outage. Instead of routing to lesser-known venues, many brokers stopped trading. This served as a wake-up call to market structure immaturity and brought attention to subpar liquidity discovery technologies.
If the fragmentation is to be controlled, time is of the essence. Many firms - in Europe and in the US - are turning to Smart Order Routing Technologies (SORT) to solve the problem, but these potential solutions come with their own set of issues. It will be interesting to see whether current fragmentation trends portend even more fragmentation in the future or if the backlash will cause another round of consolidation. Just a few years ago, many believed that the international markets would remain an oligopoly (or even a duopoly in the US), and yet we've seen the pendulum swing in the opposite direction several times before.