by George Bollenbacher
Through a Glass Darkly – First Indicators on SEF Trading
G. M. Bollenbacher & Co., Ltd.
Now that we’ve had three weeks of non-mandatory trading on SEFs, as well as the first applications to make SEF trading mandatory (the so-called “MAT” applications), can we see any early indications of how the SEF market will play out? As they say, the returns are fragmentary, but already some trends are visible.
Let’s start with the raw volume numbers. The best compilation is put together by Tod Skarecky, SVP of Clarus Financial Technology. His table for the third week of trading, 10/14-10/18, is below.
As Tod points out in his commentary, Bloomberg is the early winner in the credit space, and the telephone interdealer brokers – ICAP, Tullet, and BGC – are the early winners in rates and FX. But there is another interesting fact: the two SEFs who made the MAT applications, Javelin and trueEX, had no trading volume at all.
Another thing that is more apparent in a graph than a table is the high volatility of the volume numbers. Here’s the graph for the largest volume SEFs for the last two weeks.
We can’t tell from Tod’s data how many transactions these volumes represent, but we can from the individual SEFs’ reporting, and the USD 291 billion in volume for ICAP on 10/16 comprised 113 transactions, in a wide variety of currencies, of which 43 were larger than 1 billion, and almost all of those were FRAs. According to Tod, the spike represents a weekly fixing event. But it does raise the obvious question of whether the block thresholds set by the CFTC will make any difference at all, since these very large trades were voluntarily done on a SEF.
So what can we discern from these very early results?
1. The power in the market has already started shifting from the dealers to the customers. Given that the dealers hate SEFs because they narrow the spreads and expose trades to the daylight, the fact that any transactions are done on them voluntarily means that the dealers are losing their control of the market. Of course, as Tod points out, the phone brokers may be routing some of their phone business to their own SEFs, but the trend is evident just the same.
2. A corollary is the shift from principal to agency transactions, which will accelerate when the first MAT rulings take effect. Both trading and sales desks at SDs have been slow to come to grips with this shift, and the buy side has probably not fully comprehended it either, but everyone does business differently in an agency model. Will FCMs become the dominant brokers in the SEF world, because they will be able to supply certainty of clearing? Will the buy side be cognizant of and understand the new relationships between the brokers and the SEFs? Will the principal trade become extinct because of the CFTC’s 15-second rule?
3. It may already be too late for the marginal SEFs in the high volume products. The first mover advantage and the control of some orders may already have doomed those markets that don’t have either. The important question is, what will the marginal SEFs do about it. Two obvious possibilities: 1) ramp up the rebate game and 2) go after lower volume products. I expect both to start very soon.
4. The traditional ECNs (MarketAxess and TradeWeb, for example) appear to be late to the party, and will have to scramble to catch up. It may be that the principal-based ECN model that they have perfected isn’t well enough suited to the SEF structure, or maybe they are just getting started. In any event, they have as much to lose in this market shift as anyone, so we will watch their efforts with interest.
Of course, two weeks doesn’t tell much of a story, and this glass is still pretty foggy. But the stakes in this race are enormous, and even if the horses are just out of the gate, some front runners and some laggards are becoming visible. It promises to be an exciting six months or so.