SEF 101: Deconstructing the Swap Execution Facility

By | July 22, 2010

Also posted at and submitted to the CFTC.

The term “Swap Execution Facility” (SEF) has been formally christened and will henceforth be a part of OTC derivatives vernacular.  Yet to be determined, however, is whether SEF will become merely a new label for existing businesses or conversely an open door for competition, innovation and transparency in OTC derivatives price discovery and execution.

The Dodd-Frank financial reform bill lays the foundation defining an SEF to be:

“…a facility, trading system or platform in which multiple participants have the ability to execute or trade swaps by accepting bids and offers made by other participants that are open to multiple participants in the facility or system , through any means of interstate commerce …”

If the word “swaps” was replaced with “goods” this could be a description of eBay.  The vague definition was not accidental, however.  Congress quickly realized that trying to legislate execution methods (fully electronic, hybrid, etc.) and price discovery mechanics (continuous two-sided quotes vs. request for quote (RFQ)) is highly contentious and more importantly not critical to the stated goal of reducing systemic risk.  How exactly trades are done is of minimal consequence as long as trades are reported in a timely manner (more on what “timely” means shortly).  As I have discussed in several previous research notes, electronification of this market does not need mandating – it will come about organically under new rules.  SEFs therefore must be defined in such a way that they provide regulators with oversight of standard swap executions to ensure a fair and orderly market.

It is also important to note that SEFs are not exchanges.  If they were exchanges Congress would not have made the designation between SEFs and boards of trade (BOT).  BOTs are registered exchanges that can define contract terms,“liquidate or transfer open positions,” “require market participants in any contract to meet special margin requirements” and are required to comply with numerous other regulations.  The SEF to BOT comparison is similar to the ATS (alternative trading system) to exchange comparison in the equities world.  SEFs and ATSs are simply mechanisms for linking buyers and sellers, whereas exchanges have much broader mandates.

As we move into the rule writing process regulators do not have the Congressional luxury of being vague.  They must define with excruciating detail what an SEF is and is not.  Those details will include registration and ownership requirements, a list of what products must trade via an SEF as well as how SEFs can and must gain access to clearinghouses.  All will have a huge impact on how this market develops and whether innovation and competitive forces will overtake the status quo.

Registration & Ownership

The regulators have considerable latitude in determining how onerous (or not) the process will be for becoming an SEF.  The Dodd-Frank language states that hopefully SEFs must meet “any requirement that the Commission may impose by rule or regulation.” This is arguably more vague than the definition of SEF as discussed above.  The final requirements here are critical, however, as a filing process that is too complex would raise barriers to entry potentially eliminating innovative, would-be SEF startups and in doing so favor big established players.

The next variable in that equation is ownership limits.  The Lynch Amendment, which as part of the House Bill passed in December 2009, was set to limit dealer ownership of both execution facilities and clearinghouses.  The amendment did not become part of the final Dodd-Frank bill; however, the bill does state that regulators“shall adopt rules which may include numerical limits on the control of, or the voting rights with respect to, any derivatives clearing organization that clears swaps, or swap execution facility or board of trade designated as a contract market that posts swaps or makes swaps available for trading.” That means the spirit of the Lynch Amendment is still alive and well.

“Numerical limits” would be bad for banks (“with total consolidated assets of $50 billion or more”) and for the clearinghouses with considerable bank ownership:  LCH.Clearnet and ICE Trust, for example.  Although it remains highly improbable that anyone will be asked to divest their stake in an existing relationship, changes in the existing capital structure would likely trigger the new rules and upstart execution, or clearing venues might find it tough to find both funding and backing as both have historical come from the large dealers.

For those already in or looking to enter the OTC derivative execution business, limiting bank ownership could be huge.  If the major dealers (who are the major sources of liquidity) cannot create their own SEFs then their trade volume will have nowhere else to go but through an independent SEF.  Furthermore, with the behemoths of OTC derivatives locked out of the SEF picture the inter-dealer brokers and independent platforms will be large and in charge.  Those winners may very well still be decided based on the chosen few those same major dealers decide to trade with; however, with limited ownership in any one platform liquidity would likely be much more transient based on fees and technology.


The next question is what will trade via SEFs.  Again to the regulators – whatever they determine must be cleared must also be traded through an SEF (or Board of Trade, but I’m going to ignore that part for now).  So we can easily assume index CDS contracts, vanilla interest rate swaps (2yr, 5yr, 7yr, etc.) and potentially some FX swaps will begin trading via SEFs. TABB Group estimates show that up to 80% of interest rate swaps could be deemed standard in the next several years, ensuring their movement onto SEFs, for example.

Per the above, this does not mean execution will become immediately all electronic for these products, it only tells us that trades will be processed by registered SEFs and ultimately be reported to regulators.  For dealer-to-dealer trading this will not be a huge change, as most of that flow already goes through interdealer brokers soon to be partially rebranded as SEFs.  According to DTCC data, 83% of CDS transactions are D2D and 99.99% of transactions have a dealer on at least one side of the trade (see Exhibit 1).  For Dealer to Client (D2C) flow where trading is often done without an intermediary,  these new rules could be a boon for existing platforms such as Bloomberg or TradeWeb and new entrants looking to facilitate client flow.


Last but not least is reporting.  As stated above it’s not the method of execution that makes the SEF designation important, but the fact that trades will be reported to regulators and over time to the broader market.  The legislation states that “data relating to a swap transaction, including price and volume” must be reported in real time.  Wow – from limited transparency to real-time reporting?  Maybe.  The legislation goes on to define real time to be “as soon as technologically practicable after the time at which the swap transaction has been executed.”

The “technologically practicable” label is pretty straightforward in my eyes.  Trades should be reported within milliseconds of completion, as technology clearly makes that feasible.  However, the question here is not abouttechnology, but about when the trade is considered done.  OTC derivative markets don’t operate like equity markets where executions are finite and fast.  Hitting an offer to sell 100 shares creates a completed order of 100 shares.  For many OTC derivative transactions, accepting an initial offer to sell a $50 million IR swap does not necessarily mean the trade is over.  Further negotiating can go on between the counterparties to increase the trade size, negotiate affiliated hedge pricing or notional amounts.  Only after this process completes and the trade details affirmed can the trade be considered done.

The CFTC should not focus on “technologically practicable” but instead better define “the time at which the swap transaction has been executed.”

Moving Forward

Although competition of independent execution facilities has proven over time to decrease execution costs dramatically (OTC execution fees are much lower that typical exchange-traded fees), too much competition between clearinghouses can decrease their effectiveness.  The benefits of cross-margining and capital efficiencies that clearinghouses can provide become diluted as more clearinghouses are introduced.  A dozen clearinghouses are not good for the market.

New venues are already being born such as the recently announced Eris and Javelin platforms while existing players such as BGC, GFI, ICAP, Tradeweb and others are have begun their assault on the new world.  Whoever manages to win this new fight for liquidity, the birth of the SEF will finally bring OTC derivatives into the 21st century, driving them to utilize the technological and market structure innovations seen over the past two decades while not killing the benefits of the OTC model.  Exactly how earth shattering these changes will be largely comes down to the regulatory rule writing that is about to begin.

One thought on “SEF 101: Deconstructing the Swap Execution Facility

  1. Pingback: October 2nd Should Be Boring (Unless the CFTC Kills Electronic Trading) | Kevin on the Street

Have something to say about what I wrote? Please comment...