Week 2 of the new order in OTC derivatives execution is behind us. It’s been labeled by the media as anything from a false start, to a whimper, to a huge success. Of course there are no mandatory trades made available for trading, so the results should probably be construed as interesting at best. It’s akin to the preseason in professional sports; it can be interesting to watch, and you might even be able to discern who has got all the right stuff, but no matter how well or poorly you play, it just doesn’t matter quite yet.
THE DATA
With that said, below are the statistics I have gathered for week 2, which is the first full week of SEF activity. (For those newcomers, the week 1 summary can be found here.)
THE FINE PRINT
Again I need to share some caveats with this data:
- Data above is USD notionals, for USD based transactions.
- While I have gathered CMD and EQD data, the numbers really are insignificant to show here. I literally counted 3 entries for EQD.
- Some cells marked “#” are because the data was not available, be it the SEF or my fault in gathering.
- BGC began reporting this week, however only day’s worth is available, and is particularly difficult to interpret. I will give them some more time to present it in a more readily consumable format.
- The yellow cells represent data that was not published as of writing (early morning Monday UTC).
STANDARDS
As I mentioned in last weeks post, there is no standard format for this reporting. I went out of my way in an earlier report to congratulate BGC on their exceptionally transparent CFTC filings. However it seems the SEF data reporting group didn’t receive that memo. While it’s easy for me to be critical, we must remember that there is no explicit CFTC guidance on such reporting, so having worked in Financial Technology for many years, I can appreciate the dilemma the IT groups would have been faced with when piecing the reports together.
One such lack of standards is when to submit today’s activity. Roughly half of the SEF’s release their data and simply overwrite their previous public information, so if you didn’t get it by the “End of Day”, its gone. Which raises the next problem – when is “End of Day”? Having an FX background, I tend to recite the rule that a trading day ends at 5pm NY (Except if you are talking about New Zealand, etc). Like I say, however, we are at early Monday UTC and 7 of the SEF’s have not published Friday’s data yet. So I find myself checking early in the morning, early afternoon, and in the evening, to make sure I catch all the reports.
MARKET SHARE
The table above gives a decent glimpse of the market share per SEF, per asset class. However each asset class can be broken down further. For example, FXD can be NDFs or options, CRD can include more than Indexes and Single Names, and IRD can be swaps, FRAs, Swaptions, etc. I was curious to look at some very liquid products and see who has the market. In particular I wanted to look at 5, 10 and 30 year vanilla interest rate swaps.
Further, I suspect that as clients begin to trade more actively on SEF’s, trade sizes should tend to decrease, hence it will be interesting to see not just market share as a % of notional, but also as a trade count share.
Hence I brought up the Clarus SDRView and viewed the trade counts and notional for these liquid 5, 10 and 30 year swaps which can be seen on the table in the screenshot below.
I then plotted the data that I had gathered against this data. You’ll notice that only 2 SEF’s publish trade IRD counts (Tullet and Tradition), however it is interesting to see that the trade count % share roughly matches the notional % share. I would think most of the SEF’s are using the SDRView to do this math on trade count and notional for themselves.
WHAT NEXT?
So the game is on, we’re in the preseason, and we’re starting to see things flesh out. But when does the regular season begin?
UPDATE
This weekly issue of SEF updates is not the most current. To see all SEF posts, including the most recent, please click through to the SEF Category.