Streetwise Professor

September 26, 2011

An Unintended Consequence?

Filed under: Clearing,Commodities,Derivatives,Economics,Exchanges,Politics,Regulation — The Professor @ 6:34 pm

No, it’s not all Kudrin all the time here at SWP.  I know there are those of you looking for your clearing fix.

A brief thought about an unintended destination on the Road to SEFdom–a vertical silo.

Long time readers will know that my transactions cost economics-based analysis predicts that dominant execution venues will vertically integrate into of post-trade services, including clearing and settlement.  Many dominant markets are already “silos” and others are moving in that direction.  The days of an execution venue obtaining post-trade services from a “horizontal” CCP or settlement provider (like LCH.Clearnet) are numbered except when regulation (like RegNMS) socializes order flow.

Horizontal, independent clearers (or settlement providers) can survive in an OTC market, where there is no centralized execution venue.  Vertical integration economizes on transactions costs and double-marginalization costs when there are back-to-back monopolies (and near monopolies).  No execution venue, no back-to-back monopoly problem.

That’s where the SEF mandate comes in.  By forcing exchange-like execution of swap transactions, there is a very real possibility that a dominant execution venue will emerge–and that the resulting frictions in the dealings between that venue and its clearer will lead to a merger between them (or some sort of long-term, exclusive contract.)

Whether this outcome occurs depends on the rules for SEFs.  There is likely to be a proliferation of SEFs early on, but rules for SEFs will determine whether a handful or one survive for a particular instrument class.  If SEF regulations (a) require them to operate auctions (rather than, say, RFQs), and (b) do not impose obligations on SEFs to direct orders to other SEFs offering better prices, then network effects will lead to tipping to a single venue.  Conversely, if there are SEF order handling rules analogous to RegNMS, multiple SEFs may survive.  Under the first alternative, integration is likely to occur.  Under the second, the likely outcome is for multiple SEFs to be served by one horizontal CCP.

SEF rules are still up in the air, and I have not seen any discussion of inter-SEF obligations–or lack thereof.  In contrast to SEC, CFTC has no tradition of even contemplating, let alone imposing, order handling rules.  The issue has never really come up.  I think the presumption in the agency is there will be rigorous competition among SEFs, but that conclusion depends crucially on SEF rules.  In particular, rules relating to the way SEFs compete.  If SEFs are required to operate auction-like mechanisms, and are under no obligation to direct orders to markets displaying better prices, markets will tip to a single SEF.  This will lead to integration.  If regulators impose order handling obligations, multiple SEFs will survive, and integration is far less likely.

Has anybody thought this through?  Are order handling rules even on the table?

Print Friendly, PDF & Email


  1. @STREETWISEPROFESSOR In contrast, MIFIR in Europe does consider this aspect at least in relation to Organised Trading Facilities [“OTF”] in that an OTF
    – has a best execution obligation
    – may not integrate with other OTFs but in meeting the former obligation can integrate with Regulated Markets and Multi-lateral Trading [“MTF”] venues

    Hence, an OTF should seek out the best price including those available on other venues. However, the detail is a little more complex [isn’t it always] in that an OTF will only have to consider venues it explicitly names in its’ own Best Execution Policy, which will be part of its’ terms of usage. Whilst an OTF will be required to publish the other top 5 venues it used annually, it could of course included no others in its BestEx policy and theoretically still be compliant.

    Note – the above is based on the leaked MIFIR draft and a) may be omitted from the issued version b) get negotiated out

    Comment by John Wilson — September 27, 2011 @ 2:32 am

  2. […] More here Share this:EmailDiggStumbleUponTwitterRedditFacebookPrintLike this:LikeBe the first to like this post. […]

    Pingback by SEF Order Handling Rules « SerenDiPity — September 27, 2011 @ 5:32 am

  3. @John–thanks for that. The details of the obligation are very important. In the US, there was the Intermarket Trading System for years, and a theoretical obligation on exchanges to use it to route orders to markets showing better prices. But the system was so creaky, and the obligation so weak, that it had virtually no impact. RegNMS and the movement of the market to electronic trading socialized order flow far more effectively than ITS, and the market changed dramatically as a result, with NYSE market share falling from the mid-80s to the 30 percent range.

    I also wonder how this would work in derivatives given the potential that two SEFs could have slightly different products. There is also the whole clearing issue. The practical issues are enough to make my head hurt.

    The ProfessorComment by The Professor — September 27, 2011 @ 9:41 am

  4. When considering the future wrt SEFs I (rightly or wrongly) consider the FX Spot market as potentially a good indication of what’s to come. In FX Spot you have multiple ECNs trading essentially the same product, and smart order routing systems used by trading parties to find and execute the best price across venues. There is some differentiation by CCY pair liquidity but even that differentiator has been minimized by algos that banks and clients are running to capitalize on price differences.

    I’m not aware of any recent consolidation in the industry. On the new participant side there was a ill-fated attempt by FXMarketspace a few years ago, and some talk of a dealer only platform breaking off from EBS that was recently scuttled. Movements to algo trading have potentially helped create liquidity in secondary venues in spits of the lack of order-routing rules.

    I agree that establishment of dominant execution venues tips the scales for vertically integrated clearing but I wonder if that’s what the future has in store. If dominant execution platforms appear in the future, keep in mind that clearing, like any post trade flow tends to be quite sticky – there would need to be a really good reason for a firm to make a change.

    Just my thoughts.. will certainly be interesting to see how this all plays out 🙂

    Comment by Rogers Hornsby — October 7, 2011 @ 10:45 am

RSS feed for comments on this post. TrackBack URI

Leave a comment

Powered by WordPress