In the past I’ve spoken about buy side incentives for using an agency model – chief among them the reluctance or inability to navigate the slew of legal documents an institutional investor would need to sign in order to gain access to the needed SEFs. (Other incentives are discussed in a good post on the topic here.) What’s changed since the November 14th guidance is the last part – the number of needed SEFs. On November 13th, buy side firms trading index CDS needed access to three SEFs. Post-October 15th they need access to at least 8 to ensure they find the true best price. This also reintroduces and makes more real the idea of SEF aggregation. So while direct access is nice, the number of asset managers and hedge funds with the budget, legal, and technology staff to get this done is relatively small.
But despite the buy side’s new incentive to trade via an agency model, they don’t have to. Those small few with the budget and staff to connect directly are excited to gain access where they never could before. Furthermore, the impartial access rule applies to the SEFs, not to the FCMs. So while SEFs need to allow access for all, FCMs can use their own discretion when deciding to which SEFs they want to provide clients access. This does leave a lot of the power with the dealers, however their incentives to limit access are small. The FCM side of the business makes money on volume and notional outstanding, not on the spread. So the more chances they give their clients to trade the better. And competition now exists; either via a relatively long list of emerging FCMs or by connecting directly (see slew of legal documents required above).
SEFs have new incentive to support dealer agency offerings now as well. Many of the incumbents have dragged their feet on providing API access to their platforms, instead encouraging users to trade via their native screens as they have for years. I get this approach – like many technologies that allow API access, the richest way to interact with the underlying technology is directly through their front end (think Facebook vs. social media aggregators). But that logic starts to fall apart when competition grows. Per my above credit market example, investors won’t settle for 8 different trading screens like they did with equity markets in the 90s, they’ll want one (or maybe two). The Facebook analogy holds as well – now that I look at a half a dozen social media networks an aggregator is the only way to go. Not an aggregator that shows all posts (or available swap prices) in a single stream, but an aggregator that lets me view and post to all of those networks as I see fit. I know the analogy is an over simplification, but the parallels make a complicated story much easier to understand.
All of this still leaves the cleared swaps market in a pretty unique market structure position compared to other cleared markets: many SEFs to many CCPs, clients can connect directly or via a broker, trade execution can occur in numerous ways, etc. This is the first “new” market created after both the e-trading boom of the 2000s and the credit crisis, and despite some flaws it does seem we’re using lessons learned to make it work.