So the presidents, prime ministers and other national leaders of the European Union got together in early October, in Paris, for a summit on the banking crisis. They agreed that they should act together rather than separately to prop up the EU's banks. Then they all went home and did their own thing. Even Germany intervened to save Hypo Real Estate, after a bank-led rescue attempt collapsed. When Fortis went down, the Dutch governments dived straight in and nationalised the bank's Dutch operations, while the Belgian and Luxembourgeois governments took minority stakes in BNP Paribas, and BNP Paribas, in turn, took controlling stakes in Fortis' Belgian and Luxembourgeois operations. The Greek, Austrian and Danish governments then issued deposit guarantees. In Paris, everybody had agreed that the Irish had been wrong to get in first with their deposit guarantee.
The events of October may have been traumatic, but there is at least one positive outcome. The weeks between the Paris summit and the Gordon Brown-led triumphant-but-not-quite-successful co-ordinated mid-month rescue package for Europe (and beyond) do at least give us a clear indication of just what it takes to bridge the gap between words and actions in the official mind.
We're here to discuss MiFID, the Markets in Financial Instruments Directive, successor to the Investment Services Directive (ISD), and the first question is whether, post-October, MiFID is: (i) a step in the "wrong" direction that will be reversed because "the rules have changed" in global finance; or (ii) a step in the "wrong" direction, but don't worry, it would take a catastrophe before the authorities did anything more than call a summit about it; or (iii) a more or less welcome evolution in European regulation that has already had a broadly positive impact on auto/algo trading? And if the answer is close enough to (iii) to offset (ii) and possibly even (i), what's the new opportunity that MiFID has delivered, and can we expect more of the same in future?
To start with the rhetoric, the key phrase is "maximum harmonisation". MiFID is the Directive that requires European regulatory authorities to "harmonise", to the maximum extent possible, their regulatory regimes for investment services, and thus to, er, act in harmony to address such crises as might arise. If a finance house goes down in, say, Germany, regulators across the EU should, er, set aside their narrow national interests and work together to preserve that essential euro-harmony.
In such a crisis, the Irish government should support the Germans by not, let's say, putting their own people first.
Fine-sounding word, "harmonisation". But not one that crops up much in discussions of MiFID. The more popular term used to characterise the post-MiFID trading environment in the EU is "fragmentation". We have dark pools, dark liquidity, proliferating platforms and exchanges, much more emphasis on electronic trading solutions, not to mention an evolving definition of best practice based as much on smart-order routing as price, and then we have a whole new range of challenges around post-trade analysis, and all of it underpinned by a regulatory framework that, well, let it all happen in the space of a year. Post-MiFID Europe is either the last refuge of light-touch (but let's not say "laissez-faire") financial regulation, or it presages an era in which harmony gives way to fragmentation just as surely as the Paris summit's consensus fragmented in the face of immediate reality back home. But if fragmentation is the new harmonisation, should we worry that somebody up there is going to slam on the official brakes?
Not necessarily. First, as Bradley Duke, Head of Institutional Electronic Sales in Europe at Knight Capital, says, there's a precedent to suggest that Europe might be on the brink of cyclical fragmentation. Duke says: "In the US, fragmentation has been going for some time now. There have been several waves of framentation and consolidation already in the US. Here in Europe we're on the first wave." The US experience has a number of implications for Europe. Duke continues: "The buyside's need for new technology as well as for trading expertise in illiquid and difficult-to-trade names has increased their reliance on the sellside. There especially is demand for U.S. routing and aggregation technology expertise." And that's an irony worth exploring. But first, the second reason to believe that the post-MiFID trading environment will endure, is that the "fragmentation is harmonisation" definition just about works.
Eli Lederman, chief executive of Turquoise, says, "In the course of one week after our roll-out on 15th August, we expanded our marketplace to cover thirteen countries. I would describe that as a significant achievement made possible by MiFID." A big claim. Lederman substantiates it: "We had one regulatory process and then we were passported around. Similarly, it is possible to have members in different countries who have availed themselves of pan-European trading." Roll credits and play the happy-ending music? Not yet. Asked whether anything has surprised him since MiFID's unveiling on 1st November 2007, Lederman comments: "There are more MTFs on the horizon than we might have thought, and probably an objective observer would conclude that there are more MTFs on the horizon than really make sense."
Lederman's view is that the new MTFs will have to find ways to differentiate themselves. Prompted by a similar question, Martin Graham, Director of Equity Markets for the London Stock Exchange Group (LSE), raises the prospect of competition. Graham says: "We anticipated that it would take a while for the new landscape to take shape, and that increased competition would not prove to be a zero sum game. So far that seems to be the picture that is emerging. But what changes MiFID might have brought in the area of competition is still not really clear. In our view, competition can stimulate and create new liquidity, generating opportunities for all participants and ultimately greater market efficiency."
If harmonisation means a level playing field populated by trading venues with pan-European coverage, and if fragmentation means that they're all going to be battling for market share, then MiFID has certainly given us harmonised fragmentation, or possibly fragmented harmonisation, and if the US precedent applies, it will prove cyclical. In practical terms, it has also thrown out some new trading opportunities. Graham adds, "Certainly the extent to which one of the most active participants on Chi-X has increased the business it is doing on SETS is striking; seeking out arbitrage or hedging positions. These sorts of strategies are only possible when the speed of the platform gives certainty of execution." Spookily enough, Graham's further discussion of new initiatives at the LSE coincides with the arrival of news from Chi-X. Graham was saying: "On the technology side, this autumn sees the migration of Italian equities to our TradElect platform to create Europe's biggest single liquidity pool, and we will also increase capacity on TradElect to 20,000 continuous messages per second, and reduce latency to 3 milliseconds." While Chi-X was simultaneously emailing in with the news that it's "now successfully trading, clearing and settling five of the top component stocks of the Italian S&P MIB40 index, and planning to introduce the remainder of the index's components later in the month. The first Italian stocks available for trading on Chi-X Europe are: Enel; ENI; Fiat; Intesa Sanpaolo; UniCredit."
[For the record, the release is dated October; see www.automated trader.net for the follow-up story, as this issue goes to press at end-October.]
It's a competitive business, talking to Automated Trader. Peter Randall, CEO of Chi-X, also has competition in mind, although his main point is that MiFID has enabled pricing efficiencies. Randall says: "The idea of best execution has been liberated. Previously the understanding of best execution was between very narrow tramlines: the price on the stock exchange at the time of dealing. Now, there's a much more liberal and desirable climate where people are able to make a choice." As Graham might put it: Chi-X, SETS, or both. And we could add Turquoise and others to the list. Randall goes on to make a thought-provoking point that has a competitive sting in the tail. Randall says: "That choice has done two things. It has shocked the incumbents. So much of their volume could just disappear. It has shocked the participants who realise that a share bought on Chi-X is as good as a share bought on an incumbent exchange." But doesn't the proliferation of new trading venues introduce a new potential for pricing inefficiencies, rather than efficiencies? Randall continues, looking to the future: "We've reached the stable state, in the sense that people recognise change is both tangible and desirable. This will develop. We'll have some new entrants into the steady state but gradually, given the fact that the LSE has 1,100 people working for it and we have 27 people working for us, the economics will start to do their corrosive work."
Hmmm. Right to reply? Discussing the LSE's response to the post-MiFID launching of alternative trading venues in a variety of locations across Europe, Graham says: "One aspect of our response will be our Baikal pan-European dark pool for equities trading. Baikal will be the first true liquidity aggregator, offering smart order routing to other 'lit' and dark venues across Europe, and as such we anticipate that it will play a crucial role in improving market operation and liquidity across Europe. The platform's matching engine itself will offer a unique combination of advantages. It will be a neutral platform backed by customisable liquidity seeking strategies and sophisticated anti-gaming technology, all of which will offer buyside and sellside users the best mechanism to execute larger trades truly anonymously and with maximum efficiency."
And of course this debate could run and run. John Owens, Vice President of ECNs and Exchanges at Transaction Network Services (TNS), says: "Some of the second-tier players around Europe now view the new MTFs as a cheaper entry-level option to trade London stocks without having to become a member of the LSE." And Randall concedes: "You've got to believe that the stock exchanges are capable of reinventing themselves."
Indeed. Choice and competition this year; and given that this is a cyclical process, who's for consolidation next year? Or is that overly pessimistic? There's some evidence to suggest that MiFID is enlarging the market. Owens continues: "We're seeing a good number of enquiries, from new and existing customers alike, in mainland Europe, who are talking about trading London stocks, not the LSE but London stocks, because that's something that they're able to do." So there's a potential source of new liquidity.
And the US precedent is another spur to new entrants. Chris Smith, Head of Business Development, Euro Millennium and NYFIX International Director, says: "MiFID was the starting gun for a revolution in trading in Europe.
We saw MiFID as the catalyst that created the opportunity for us." Hence the arrival of the NYFIX Euro Millennium neutral dark pool. Smith says: "One thing that has surprised me is the sheer volume of order flow that goes potentially unexecuted in a day." That is clearly an opportunity, but does mopping up such "leftovers" account for the whole of the opportunity? No. Smith says: "What we're looking at here is either a new means of fulfilling the unexecuted liquidity that's always been there, or we're looking at new liquidity coming to the market because of the new means of trading and execution, light and dark, or we're looking at both. I think we're looking at both."
So. New means both enlarging the market and cutting unfilled trades. But we still haven't quite reached the happy ending. The idea of best execution may have been liberated, and new trading opportunities, and players, may have begun to appear in the newly competitive landscape, but MiFID's first year hasn't passed without introducing some new complexities. Developing his earlier point about routing and aggregation technology, Duke says: "The execution imperative is not as easy to achieve as it was before.
The most important thing in the marketplace is smart order routing. A year ago, everybody was talking about different algorithms and different benchmarks, but at the end of the day, all those algorithms need to be tuned to a fragmentated marketplace, or they simply won't work. As more players join in, and the market gets more fragmented, they'll work even less well." Richard Balarkas, CEO of Instinet, even goes so far as to say: "Smart order routing technology is the key differentiating feature between brokers who should now close their business and brokers who are positioned to take advantage of the market going forward."
Discussing the fragmentation of the market, Simon Nathanson, CEO of Neonet, says: "It's one thing to set up best execution policies, categorise clients and do the transaction reporting. That is what most people have done prior to MiFID, but not so many are prepared to trade on a fragmented market, to find the best price and to use the tools you need to use, to execute in the best way. " Like Smith, Nathanson describes MiFID as an opportunity, and speaks of Neonet's response in terms of technology, and specifically smart order routing.
That is not just a matter of achieving best execution, but of proving it. Nathanson says: "You have to look at it all the time and see how well we did the execution. Was it good enough, what price improvement did we achieve? That will become more important over time." The underlying problem being that the simple benchmarking question - Could I have done better elsewhere? - becomes significantly more difficult to answer in a fragmented market. On a similar point, Randall says: "As a result of MiFID, there is now a plethora of different reporting venues, and it becomes quite difficult for participants in the market to be able to do the analysis because they've got to go to so many places."
One answer, Duke suggests, would be a further iteration of MiFID explicitly taking up US solutions such as a consolidated tape. The snag is not just that best execution has become complex in itself, but that this complexity has a knock-on effect down the settlement chain. And MiFID doesn't provide much guidance, either. Bertrand Rassat, Managing Director of Flextrade's London office, says: "The Directive is vague in describing how best execution could and should be achieved. It's down to each asset manager's interpretation." In a Europe of multiple competing MTFs, dark and light pools, best execution is necessarily a matter of interpretation, or of policy negotiation between consenting adults.
As Rassat says: "There will be a big emphasis on post-trade analysis, because MiFID forces the buyside to quantify best execution. You will see post-trade analysis becoming more and more critical, because that's the way for the buyside traders to assess the performance of their brokers."
Should we worry? Lee Hodgkinson, CEO, SWX Europe, says: "Post-trade data has fragmented far more than pre-trade. That is a concern that we're hearing more and more from institutional customers. A full-service investment bank will execute perhaps on the exchange, they'll put some in a dark pool, some in their own dark pool, they'll sell some to another client, they'll do some algo-ing, and you've got to collate all of that data. Then the question is, how does the institution know that's best?" This is, as Hodgkinson says, the big post-trade question of today. But here's the sting in the tail. Post-trade analysis may indeed require granular data not only on the fragmented trade itself but on the various alternative trades-not-taken. But as new and established trading venues develop a track record for post-MiFID execution, best execution is itself developing a data requirement. John Coulter, Vice President of Corporate Strategy at Vhayu, says: "You have to be aware of prices in all these market centres. People are rushing to get some kind of cross-linkage together. If you don't have the data, you're not going to trade."
There's a MiFID requirement that if you're doing any internal crossing, you have to store a year's worth of quote data. But this is an active data requirement rather than just a storage issue. Just as fragmentation is the new harmonisation, so post-trade data is the new pre-trade necessity.
Coulter says: "You have to be able to quote in all these different markets now. You have to be aware of prices in all these market centres. The fragmentation has lent itself to an increased usage of computerised algorithms to find best price."
And for a magazine like this one, that counts as a happy ending.