March 6, 2013 
The CFTC had a large meeting to discuss futurised OTC products, below is a 15 minute extract of the full 6 hours where Kim Taylor (President of their Clearing Business) and Dan Maguire (President of Yorkshire, and a Senior in SwapClear US), speak about the topic.
Key points from Kim
- Capital was insufficient to cover market losses in the crisis, leading to a spiral of margin calls and a bail out, Dodd Frank is the US approach to solving this problem
- The DFA is one solution, to redesign the OTC market, and enhanced use of Futures is also a legitimate approach to the same goals
- There are 40-50 ‘line items’ for Deliverable Swap Futures, meaning 40-50 distinct products
- There were 60,000 ‘line items’ for Lehman in SwapClear. BH: I don’t count each slight permutation of an IRS as a distinct “product”, so an IRS based on ACT/360 isn’t so different from a swap based on ACT/365, in my opinion
- The diverse set of line items in the OTC market prevents liquidity formation. BH: I disagree, true liquidity can be measured from market activity by looking inside MarkitWire / SwapClear, and then into the CME Deliverable Swap Future stats.
- Compare turnover of trades to open positions, in futures the market rotates 25 times per year, in OTCs about 2.5 times (no background data provided)
- There are an estimated 5m participants in the futures market, compared to 30,000 in the OTC market.
- BH comments:
- But do those 5m all trade deliverable swap futures? And would they step up to take the other side of a DSF when CME were trying to sell off a defaulters portfolio?
- The SwapClear model mandates that surviving members must participate in the auction process, of which there are now around 70 corporate groups representing a large proportion of the whole swap market.
- Are traders in DSFs obliged to participate in a liquidation situation?
- The reduced number of line items means trade netting is an order of magnitude better in futures than OTC
- BH: Both CME and SwapClear provide trade netting of OTC IRS in their CCPs, it’s relatively easy, although not intrinsic to the IRS product the way a future contract is
- Methods of liquidation in Deliverable Swap Futures
- Open market sale
- Private negotiated sale
- Competitive auction
- For an OTC default CME only use a competitive auction, due to the different profile of an IRS portfolio
- An FCM must post the minimum CCP margin, but can collect more if they wish
Key points from Dan
- SwapClear contains 60-70% of the global OTC IR swap market
- SwapClear has cleared $19trn of the $20trn buy-side Swaps anywhere in the world (i.e. other CCPs have only captured $1trn of buy-side business)
- SwapClear clears $2trn notional per day, higher than implied by Kim perhaps
- Also torn up $170trn of IRS (I assume he means in partnership with TriOptima / TriReduce)
- Workflow for OTC is now more standardised, the flow from execution to clearing is much smoother
- OTC products hedge non-standard risks – the raison d’être of the market
- The portfolio at SwapClear when Lehman defaulted was 66,000 IRS in 5 currencies, $9trn of notional, maximum maturity of 30 years
- 30% to 40% of Lehman’s IM was used during the Default Management Process, the remainder returned to the estate of Lehman / their administrators
- No-one in SwapClear lost a penny as a result (is that a US cent or an English 1p?)
- Rate risk can be transferred and transformed, but doesn’t “disappear” so transmuting OTC Swaps into Futures will just move the problem, not solve it
- Why would a $10m DV01 in exposure attract a 2 day VaR holding period on an Exchange, and the same risk as an OTC cleared product attract a 5 day holding period?
- The key is: do you have access to liquidity in a default?
- Shouldn’t the holding period in any market be derived from the liquidity [and other key issues in a default], rather than by top down regulatory mandate?
I cut Dan off slighty short on the video, sorry Dan.
My key points
- You can download a spreadsheet of volumes of DSFs here: http://www.cmegroup.com/trading/interest-rates/dsf-volume-and-oi-tracker.html
- More work needs to be done to relate the holding period on a VaR calculation back to the participants in the market, and conditions in a default.
- Likewise the legal basis which obliges (or doesn’t) oblige folks to take a defaulters positions need examination
- The point about DV01 above is worth more examination
Source videos, 6 hours long: http://www.cftc.gov/PressRoom/Events/opaevent_cftcstaff013113
Amir Khwaja
Ben Larah
Michael Beaton
corinaki24
Jon Skinner
Mihir Vaidya
Maria Leontiou
samihodgson
Bill Hodgson
Tom Riesack
Bill
good summary Of a complex topic.
One point to add – Dan – is surely the Duke of Wigan not president of yorkshire….
Rod
Thanks Bill. A few comments:
- Position netting comparisons between IRS and swapfutures are a matter only of operational convenience (possibly reporting transparency also). More importantly risk / margin netting works fine across swaps (and swapfutures like ERIS flexes for that matter) with different dates / interest basis as do compression mechanisms like TriReduce which have reduced line items substantially therefore.
- Whilst IM costs would be lower, if swapfutures have the same raw risk before IM as swaps, then there has to be more tail risk remaining exposed. Specifically FCMs and CCPs are bearing more tail risk on swapfutures than the economically equivalent swaps. So substitution of swapfutures as currently calibrated is increasing systemic risk in FCMs and CCPs.
- Both Basel III and CFTC futures residual interest proposals as currently proposed will introduce higher capital charges for FCMs for swapfutures than economically equivalent swaps.
- On market liquidity, IRS ECNs (e.g. TradeWeb, Bloomberg etc.) have historical IRS volumes in the $10s of trillions but SEF rules may impact these volumes, whereas swapfutures have open interest of $5-6 billion at present. Neither SEFs nor exchange traded swapfutures can be said to have proven liquidity yet. In reality market liquidity is likely to be established in response to relative trading costs.
- Currently CCPs set IM levels and block sizes for futures, whereas CFTC sets minimum standards for IM and block sizes for swaps. This differential amounts to CFTC, if it makes no adjustments, de facto picking swapfutures over swaps. This breaks one of Commissioner O’Malia’s principles in his speech on the topic.
- Setting regulatory complexity concerns aside for one moment, one logical resolution would be for CFTC to harmonize the trading costs as follows:
1. Adjust the residual interest proposal so that it is financially equivalent to LSOC even if not identical in other respects (to avoid disadvantaging futures)
2. Set consistent minimum holding periods for the same economics e.g. 2-day for IMM swaps, CME deliverable IRS futures and ERIS standards, 5-day for other swaps, ERIS flexs (to remove swap disadvantages to trading participants and swapfutures disadvantages to FCMs and CCPs)
3. Set consistent minimum block sizes across exchange-traded swapfutures and SEF-traded swaps (to remove swap disadvantages)
Then we would have a level playing and the market would pick the winners and losers without regulatory distortion.