Fixed income markets and 'self-trading'
15 July 2015
Anshuman Jaswal
The interesting insights from a recent regulatory report into the operation of the US fixed income market on October 15 2014 continue to intrigue observers and raise important questions about how the market trades. A remarkable point referred to the high proportion of 'self-trading' between mainly prop traders' own automated systems. This was 14.9% for the 10-year cash Treasuries and 11.5% for the futures equivalent. While this might not have been the leading reason for the high levels of volatility on that day, it is certainly an attribute that requires careful attention. Logically speaking, such self-trading can lead to artificial market levels and prices that are quite different from what they would be without high levels of self-trading. Price volatility can also be affected by high proportion of self-trading. Prop traders often have several independent algorithms running in the market and therefore are more liable to be affected by this phenomenon. The regulators' report also noted how other participants such as banks and hedge funds did not have much self-trading. While the jury is still out on the undesirable effects of self-trading, and how responsible it was for any price swings, we can agree that it cannot really be desirable for the markets. Some exchanges and trading firms have already taken steps to curb such trading, but it is important that the firms that are more susceptible to undertake such trades try to stamp them out on their own, even if it means slowing their trading a little for the sake of reducing lower systemic risk overall.