Tales makes an interesting point on the problem of scaling up a people intensive dealer/broker activity: sales coverage of clients. He considers the problem from the perspective of scaling up the number of clients an individual sales person can cover. In my experience there’s another angle – moving up the value chain. I see dealers attempting to strip out cost by putting more of their vanilla flow business on electronic channels: Bloomberg, TradeWeb, Caplin based single dealer platforms. In my world that flow is European Govt Bonds, US Treasuries, vanilla swaps. Autoquoting, autonegotiation and STP mean that trades are priced, executed and settled with no manual intervention. A big advance on the voice trading of 10 years ago. A few years back I was toubleshooting on the desk when I heard one of our market making traders refuse to take a client voice inquiry for a small ticket, he just said “do it on Bloomberg !”

With vanilla flow on electronic channels, sales and trading can focus on higher margin structured deals, thereby scaling up earning power. Of course constant competitive pressure pushes back. Products that were bespoke voice trades become electronic trades: for example flys, custom swaps, custom structures.

Order book resiliency

June 16, 2009

In Agressive Orders and the Resiliency of a Limit Order Market Degryse et al present an interesting study of the Paris Bourse order book reaction to incoming order types. They study how best bid/offer, spreads and depth evolve in reaction to incoming market and limit orders, and to what degree they revert. Their findings, which are based on data from 1998,  are intuitively plausible. For me the surprise came in seeing the authors note the average interval between best limit updates – in one case 84 seconds !  That seems like an eternity when compared to Eurex order books when the market is moving !