NITE's (as well as SCH's, Dempsey's, etc) market making segment has experienced quickly dropping revenue capture per share. Sure, vol has decreased, but are there other long-term forces going on to decrease their profitability? How is trading from a principal basis (without enjoying the monopolistic setup of an NYSE specialist) changing? Why not just send an order to agency based ECN who isn't going to screw with my order? Not clear to me why principal based markets maker are going to be able to hang around for long.
I work for a competitor of NITE, and I have to agree. The profits margins are WAY down, and expenses are up (due to more per ticket charges (decimals).. Plus customers penny you on prints. instead of making .05 when deciamlization first took place, customers now let you make .02-.04, sometimes .01 /share on the prints... Having said that, there will always be a niche,. For example, many buy-side firms send out baskets (too many stocks for teh buyside guy to trade at once) and we can handle this. Also, opening and closing order business is popular. The future is pretty dim though. I am exploring other options as we speak.
Program trading comprises roughly 1/3 of buy-side flow these days versus 5% in the late nineties. As this number increases and smaller order sizes change the dynamics of block trading, the "traditional" market making model may be increasingly challenged. "Algorithmic trading" seems to be the catchphrase of the day. It would be interesting to fast-forward five years and see how human sell-side traders are adding value to the marketplace. Perhaps their skillset will shift much more towards quantitative analysis and pseudo-programming.
How does an increase in program trading negatively impact the profitability of specialists & market makers (those who provide liquidity)? I've read many times it does, but can't find precisely why. Program trading has stablized to around 40% for past few quarters. What would drive this % to increase. How quick is growth rate? What are other reasons you see the liquidity providing market makers' role fading?
One perspective on this is that as traders execute more trades using anonymous algorithmic strategies, market makers and specialist lose their primary advantage (seeing the order flow). That means they have less information available to make trading decisions and their risk increases. As risk increases and reward decreases (driven by tighter spreads/smaller execution sizes) the "traditional" market maker model becomes increasingly ineffective. Program/quantitative executions are on the rise for several reasons, primarily lower cost, anonymity, and execution quality. Regarding the latter, as spreads have shrunk, displayed liquidity has shrunk. That makes it harder to execute big trades without moving the stock. So more and more traders are using algorithms to slice up their orders into smaller bits and execute throughout the day "when the time is right." One innovator in this space is ATD: http://www.automatedtradingdesk.com/ Whatever the case, market makers are a tough breed and will adapt to the new realities. Many market makers have already shifted their models to facilitate this type of flow, offsetting some or all of the decrease in their traditional business.
I've been a mm for many years, until 2003. The MM as we know it is dead. Yes there wil always be a need for a human to provide liquidity and stand ready to buy and sell, but the head count will go way down (where there was 200 traders there will be 5). The black box method will prevail and firms will save saleries as well as have less liability worrying about a rouge trader etc
Strong comments. What differences are you seeing in order flow? Where is it going to (ie; ECNs) and why? Not clear to me why there is a need for MMs except for illiquid stocks, which are largely the low priced and unprofitably traded OTC BB.
Order flow has diminshed. Over 3/4 of it (since 2 year ago) has gone to ECN's , and the order flow you DO get is purely relationship based, or in return for research.. Since ECN's are so cheap, you can only make .01-.04 /share usually. Any more, and the client gets in trouble from there superior or their client, for NOT USING an ECN the ability for a trader to "add value" in a super-soes environment is next to impossible. To answer your question, there really isnt a need anymore.. ECNs can do it all, and do it all cheaper. Us market makers are a dying breed, hanging on to something that is going away...